>>Hi, Maya. Mr England. And I am an associate professor of electrical engineering and computer science at M I T. It's been fantastic to be part of this team that Professor Yamamoto put together, uh, for the entity Fire program. It's a great pleasure to report to you are update from the first year I will talk to you today about our recent work in photonic accelerators for machine intelligence. You can already get a flavor of the kind of work that I'll be presenting from the photonic integrated circuit that services a platonic matrix processor that we are developing to try toe break some of the bottle next that we encounter in inference, machine learning tasks in particular tasks like vision, games control or language processing. This work is jointly led with Dr Ryan heavily, uh, scientists at NTT Research, and he will have a poster that you should check out. Uh, in this conference should also say that there are postdoc positions available. Um, just take a look at announcements on Q P lab at m i t dot eu. So if you look at these machine learning applications, look under the hood. You see that a common feature is that they used these artificial neural networks or a and ends where you have an input layer of, let's say, and neurons and values that is connected to the first layer of, let's Say, also and neurons and connecting the first to the second layer would, if you represented it biomatrix requiring and biomatrix that has of order and squared free parameters. >>Okay, now, in traditional machine learning inference, you would have to grab these n squared values from memory. And every time you do that, it costs quite a lot of energy. Maybe you can match, but it's still quite costly in energy, and moreover, each of the input values >>has to be multiplied by that matrix. And if you multiply an end by one vector by an end square matrix, you have to do a border and squared multiplication. Okay, now, on a digital computer, you therefore have to do a voter in secret operations and memory access, which could be quite costly. But the proposition is that on a photonic integrated circuits, perhaps we could do that matrix vector multiplication directly on the P. I C itself by encoding optical fields on sending them through a programmed program into parameter and the output them would be a product of the matrix multiplied by the input vector. And that is actually the experiment. We did, uh, demonstrating that That this is, you know, in principle, possible back in 2017 and a collaboration with Professor Marine Soldier Judge. Now, if we look a little bit more closely at the device is shown here, this consists of a silicon layer that is pattern into wave guides. We do this with foundry. This was fabricated with the opposite foundry, and many thanks to our collaborators who helped make that possible. And and this thing guides light, uh, on about of these wave guides to make these two by two transformations Maxine and the kilometers, as they called >>input to input wave guides coming in to input to output wave guides going out. And by having to phase settings here data and five, we can control any arbitrary, uh, s U two rotation. Now, if I wanna have any modes coming in and modes coming out that could be represented by an S u N unitary transformation, and that's what this kind of trip allows you to dio and That's the key ingredient that really launched us in in my group. I should at this point, acknowledge the people who have made this possible and in particular point out Leon Bernstein and Alex lots as well as, uh, Ryan heavily once more. Also, these other collaborators problems important immigrant soldier dish and, of course, to a funding in particular now three entity research funding. So why optics optics has failed many times before in building computers. But why is this different? And I think the difference is that we now you know, we're not trying to build an entirely new computer out of optics were selective in how we apply optics. We should use optics for what it's good at. And that's probably not so much from non linearity, unnecessarily I mean, not memory, um, communication and fan out great in optics. And as we just said, linear algebra, you can do in optics. Fantastic. Okay, so you should make use of these things and then combine it judiciously with electronic processing to see if you can get an advantage in the entire system out of it, okay. And eso before I move on. Actually, based on the 2017 paper, uh, to startups were created, like intelligence and like, matter and the two students from my group, Nick Harris. And they responded, uh, co started this this this jointly founded by matter. And just after, you know, after, like, about two years, they've been able to create their first, uh, device >>the first metrics. Large scale processor. This is this device has called Mars has 64 input mode. 64 Promodes and the full program ability under the hood. Okay. So because they're integrating wave guides directly with Seamus Electron ICS, they were able to get all the wiring complexity, dealt with all the feedback and so forth. And this device is now able to just process a 64 or 64 unitary majors on the sly. Okay, parameters are three wants total power consumption. Um, it has ah, late and see how long it takes for a matrix to be multiplied by a factor of less than a nanosecond. And because this device works well over a pretty large 20 gigahertz, you could put many channels that are individually at one big hurts, so you can have tens of S U two s u 65 or 64 rotations simultaneously that you could do the sort of back in the envelope. Physics gives you that per multiply accumulate. You have just tens of Tempted jewels. Attn. A moment. So that's very, very competitive. That's that's awesome. Okay, so you see, plan and potentially the breakthroughs that are enabled by photonics here And actually, more recently, they actually one thing that made it possible is very cool Eyes thes My face shifters actually have no hold power, whereas our face shifters studios double modulation. These use, uh, nano scale mechanical modulators that have no hold power. So once you program a unitary, you could just hold it there. No energy consumption added over >>time. So photonics really is on the rise in computing on demand. But once again, you have to be. You have to be careful in how you compare against a chance to find where is the game to be had. So what I've talked so far about is wait stationary photonic processing. Okay, up until here. Now what tronics has that also, but it doesn't have the benefits of the coherence of optical fields transitioning through this, uh, to this to this matrix nor the bandwidth. Okay, Eso So that's Ah, that is, I think a really exciting direction. And these companies are off and they're they're building these trips and we'll see the next couple of months how well this works. Uh, on the A different direction is to have an output stationary matrix vector multiplication. And for this I want to point to this paper we wrote with Ryan, Emily and the other team members that projects the activation functions together with the weight terms onto a detector array and by the interference of the activation function and the weight term by Hamad and >>Affection. It's possible if you think about Hamad and affection that it actually automatically produces the multiplication interference turn between two optical fields gives you the multiplication between them. And so that's what that is making use of. I wanna talk a little bit more about that approach. So we actually did a careful analysis in the P R X paper that was cited in the last >>page and that analysis of the energy consumption show that this device and principal, uh, can compute at at an energy poor multiply accumulate that is below what you could theoretically dio at room temperature using irreversible computer like like our digital computers that we use in everyday life. Um, so I want to illustrate that you can see that from this plot here, but this is showing. It's the number of neurons that you have per layer. And on the vertical axis is the energy per multiply accumulate in terms of jewels. And when we make use of the massive fan out together with this photo electric multiplication by career detection, we estimate that >>we're on this curve here. So the more right. So since our energy consumption scales us and whereas for a for a digital computer it skills and squared, we, um we gain mawr as you go to a larger matrices. So for largest matrices like matrices of >>scale 1,005,000, even with present day technology, we estimate that we would hit and energy per multiply accumulate of about a center draw. Okay, But if we look at if we imagine a photonic device that >>uses a photonic system that uses devices that have already been demonstrated individually but not packaged in large system, you know, individually in research papers, we would be on this curve here where you would very quickly dip underneath the lander, a limit which corresponds to the thermodynamic limit for doing as many bit operations that you would have to do to do the same depth of neural network as we do here. And I should say that all of these numbers were computed for this simulated >>optical neural network, um, for having the equivalent, our rate that a fully digital computer that a digital computer would have and eso equivalent in the error rate. So it's limited in the error by the model itself rather than the imperfections of the devices. Okay. And we benchmark that on the amnesty data set. So that was a theoretical work that looked at the scaling limits and show that there's great, great hope to to really gain tremendously in the energy per bit, but also in the overall latency and throughput. But you shouldn't celebrate too early. You have to really do a careful system level study comparing, uh, electronic approaches, which oftentimes happened analogous approach to the optical approaches. And we did that in the first major step in this digital optical neural network. Uh, study here, which was done together with the PNG who is an electron ICS designer who actually works on, uh, tronics based on c'mon specifically made for machine on an acceleration. And Professor Joel, member of M I t. Who is also a fellow at video And what we studied there in particular, is what if we just replaced on Lee the communication part with optics, Okay. And we looked at, you know, getting the same equivalent error rates that you would have with electronic computer. And that showed that that way should have a benefit for large neural networks, because large neural networks will require lots of communication that eventually do not fit on a single Elektronik trip anymore. At that point, you have to go longer distances, and that's where the optical connections start to win out. So for details, I would like to point to that system level study. But we're now applying more sophisticated studies like this, uh, like that simulate full system simulation to our other optical networks to really see where the benefits that we might have, where we can exploit thes now. Lastly, I want to just say What if we had known nominee Garrity's that >>were actually reversible. There were quantum coherent, in fact, and we looked at that. So supposed to have the same architectural layout. But rather than having like a sexual absorption absorption or photo detection and the electronic non linearity, which is what we've done so far, you have all optical non linearity, okay? Based, for example, on a curve medium. So suppose that we had, like, a strong enough current medium so that the output from one of these transformations can pass through it, get an intensity dependent face shift and then passes into the next layer. Okay, What we did in this case is we said okay. Suppose that you have this. You have multiple layers of these, Uh um accent of the parameter measures. Okay. These air, just like the ones that we had before. >>Um, and you want to train this to do something? So suppose that training is, for example, quantum optical state compression. Okay, you have an optical quantum optical state you'd like to see How much can I compress that to have the same quantum information in it? Okay. And we trained that to discover a efficient algorithm for that. We also trained it for reinforcement, learning for black box, quantum simulation and what? You know what is particularly interesting? Perhaps in new term for one way corner repeaters. So we said if we have a communication network that has these quantum optical neural networks stationed some distance away, you come in with an optical encoded pulse that encodes an optical cubit into many individual photons. How do I repair that multi foot on state to send them the corrected optical state out the other side? This is a one way error correcting scheme. We didn't know how to build it, but we put it as a challenge to the neural network. And we trained in, you know, in simulation we trained the neural network. How toe apply the >>weights in the Matrix transformations to perform that Andi answering actually a challenge in the field of optical quantum networks. So that gives us motivation to try to build these kinds of nonlinear narratives. And we've done a fair amount of work. Uh, in this you can see references five through seven. Here I've talked about thes programmable photonics already for the the benchmark analysis and some of the other related work. Please see Ryan's poster we have? Where? As I mentioned we where we have ongoing work in benchmarking >>optical computing assed part of the NTT program with our collaborators. Um And I think that's the main thing that I want to stay here, you know, at the end is that the exciting thing, really is that the physics tells us that there are many orders of magnitude of efficiency gains, uh, that are to be had, Uh, if we you know, if we can develop the technology to realize it. I was being conservative here with three orders of magnitude. This could be six >>orders of magnitude for larger neural networks that we may have to use and that we may want to use in the future. So the physics tells us there are there is, like, a tremendous amount of gap between where we are and where we could be and that, I think, makes this tremendously exciting >>and makes the NTT five projects so very timely. So with that, you know, thank you for your attention and I'll be happy. Thio talk about any of these topics

>>from the Cube Studios in Palo Alto and Boston connecting with thought leaders all around the world. This is a cube conversation. Alright, welcome back to help us dig into this discussion and happy to welcome to the program. Chris Prasad. He is the senior vice president and general manager of the V Sphere business And just Simon, chief technologist for the high performance computing group. Both of them with VM ware. Gentlemen, thanks so much for joining. >>Thank you for having us. >>All right, Krish. When VM Ware made the bit fusion acquisition, everybody was looking the You know what this will do for this space GP use? We're talking about things like AI and ML. So bring us up to speed. As to, you know, the news today is the what being worth doing with fusion. >>Yeah. Today we have a big announcement. I'm excited to announce that, you know, we're taking the next big step in the AI ml and more than application strategy. With the launch off bit fusion, we just now being fully integrated with the V Sphere seven black home and we'll be releasing this very shortly to the market. As you said when we acquired institution a year ago, we had a showcase that's capable base as part of the normal event. And at that time we laid out a strategy that part of our institution as the cornerstone off our capabilities in the platform in the Iot space. Since then, we have had many customers. Take a look at the technology and we have had feedback from them as well as from partners and analysts. And the feedback has been tremendous. >>Excellent. Well, Chris, what does this then mean for customers, you know, what's the value proposition? That diffusion brings the visa versa? >>Yeah, if you look at our customers, they are in the midst of a big ah journey in digital transformation. And basically, what that means is customers are building a ton of applications, and most of those applications have some kind of data analytics or machine learning embedded in it. And what this is doing is that in the harbor and infrastructure industry, this is driving a lot of innovation. So you see the admin off a lot off specialized accelerators, custom a six FPs. And of course, the views being used to accelerate the special algorithms that these ai ml type applications need And, um, unfortunately, customer environment. Most of these specialized accelerators in a bare metal kind of set up. So they're not taking advantage off optimization and everything that it brings to that. Also, with fusion launched today, we are essentially doing the accelerator space. What we need to compute several years ago. And that is, um, essentially bringing organization to the accent leaders. But we take it one step further, which is, you know, we use the customers the ability to pull these accelerators and essentially going to be a couple of from the server so you can have a pool of these accelerators sitting in the network, and customers are able to then target their workloads and share the accelerators, get better utilization, drive a lot of cost improvements and, in essence, have a smaller pool that they can use for a whole bunch of different applications across the enterprise. That is a huge enabler for our customers. And that's the tremendous positive feedback that we get getting both from customers as well. >>Excellent. Well, I'm glad we've got Josh here to dig into some of the pieces, but before we get to you they got Chris. Uh, part of this announcement is the partnership of VM Ware in Dell. So tell us about what the partnership is in the solutions for for this long. >>Yeah. We have been working with the Dell in the in the AI and ML space for a long time. We have, ah, good partnership there. This just takes the partnership to the next level, and we will have, ah, execution solution support in some of the key. I am. It'll targeted the words like the sea for 1 40 the r 7 40 Those are the centers that would be partnering with them on and providing solutions. >>Okay, Tough. Take us in a little bit further as how you know the mechanisms of diffusion work. >>Yeah, that's a great question. So think of it this way. There there is a client component that we're using in a server component. The server component is running on a machine that actually has the physical GP use installed in it. The client machine, which is running the bit fusion client software, is where the user, the data scientist, is actually running their machine machine learning application. But there's no GPU actually in that host. And what is happening with fusion technology is that it is essentially intercepting the Cuda calls that are being made by that machine learning application and promoting those protocols over to the bit fusion server and then injecting them into the local GPU on the server. So it's actually, you know, we call it into a position in the ability that remote these protocols, but it's actually much more sophisticated than that. There are a lot of underlying capabilities that are being deployed in terms of optimization who takes maximum advantage of the, uh, the networking link that's it between the client machine and the server machine. But given all of that, once we've done it with diffusion, it's now possible for the data scientist either consume multiple GP use for single GPU use or even fractional GP use across that interconnected using the using technology. >>Okay, maybe it would help illustrate some of these technologies. If you got a couple of customers. >>Yeah, sure. So one example would be a retail customer. I'm thinking of who is. Actually it's ah grocery chain that is deploying ah, large number of video cameras into their into their stores in order to do things like, um, watch for pilfering, uh, identify when storage store shelves could be restocked and even looking for cases where, for example, maybe a customer has fallen down in denial on someone needs to go and help those multiple video streams and then multiple applications that are being run that part are consuming the data from those video screens and doing analytics and ml on them would be perfectly suited for this type of environment where you would like to be ableto have these multiple independent applications running. But having them be able to efficiently share the hardware resources of the GP is another example would be retailers who are deploying ML our check out registers who helped reduce fraud customers who are buying, buying things with, uh, fake barcodes, for example. So in that case, you would not necessarily want to deploy ah single dedicated GPU for every single check out line. Instead, what you would prefer to do is have a full set of resource. Is that each inference operation that's occurring within each one of those check out lines but then consume collectively. That would be two examples of the use of this wonderful in technology. >>Okay, great. So, Josh, last question for you is this technology is this only for use and anything else? You can give us a little bit of a look forward as to what we should be expecting from the big fusion technology. >>Yeah. So currently, the target is specifically NVIDIA gpu use with Buddha. Ah, the team, actually, even prior to acquisition had done some work on enablement of PJs. And also, I have done some work on open CL, which is more open standard for device access. So what you will see over time is an expansion of the diffusion capabilities to embrace devices like F PJs of the domain. Specific. A six that was referring to earlier will roll out over time, but we are starting with the NVIDIA GPU, which totally makes sense, since that is the primary hardware acceleration. And for deep learning currently >>excellent. Well, John and Chris, thank you so much for the updates to the audience. If you're watching this live leads growing, the crowd chat out Im to ask your questions. This page, if you're watching this on demand, you can also go to crowdchat dot net slash make ai really to be able to see the conversation that we had. Thanks so much for joy. Yeah, yeah, yeah, >>yeah.

>> Announcer: From the SiliconANGLE Media office in Boston, Massachusetts, it's theCUBE. Now here's your host, Dave Vellante. >> In the early days of big data and Hadoop, the focus was really on operational efficiency where ROI was largely centered on reduction of investment. Fast forward 10 years and you're seeing a plethora of activity around machine learning, and deep learning, and artificial intelligence, and deeper business integration as a function of machine intelligence. Welcome to this Cube conversation, The Skinny on Machine Intelligence. I'm Dave Vellante and I'm excited to have Jim Kobielus here up from the District area. Jim, great to see you. Thanks for coming into the office today. >> Thanks a lot, Dave, yes great to be here in beautiful Marlboro, Massachusetts. >> Yes, so you know Jim, when you think about all the buzz words in this big data business, I have to ask you, is this just sort of same wine, new bottle when we talk about all this AI and machine intelligence stuff? >> It's actually new wine. But of course there's various bottles and they have different vintages, and much of that wine is still quite tasty, and let me just break it out for you, the skinny on machine intelligence. AI as a buzzword and as a set of practices really goes back of course to the early post-World War II era, as we know Alan Turing and the Imitation Game and so forth. There are other developers, theorists, academics in the '40s and the '50s and '60s that pioneered in this field. So we don't want to give Alan Turing too much credit, but he was clearly a mathematician who laid down the theoretical framework for much of what we now call Artificial Intelligence. But when you look at Artificial Intelligence as a ever-evolving set of practices, where it began was in an area that focused on deterministic rules, rule-driven expert systems, and that was really the state of the art of AI for a long, long time. And so you had expert systems in a variety of areas that became useful or used in business, and science, and government and so forth. Cut ahead to the turn of the millennium, we are now in the 21st century, and what's different, the new wine, is big data, larger and larger data sets that can reveal great insights, patterns, correlations that might be highly useful if you have the right statistical modeling tools and approaches to be able to surface up these patterns in an automated or semi-automated fashion. So one of the core areas is what we now call machine learning, which really is using statistical models to infer correlations, anomalies, trends, and so forth in the data itself, and machine learning, the core approach for machine learning is something called Artificial Neural Networks, which is essentially modeling a statistical model along the lines of how, at a very high level, the nervous system is made up, with neurons connected by synapses, and so forth. It's an analog in statistical modeling called a perceptron. The whole theoretical framework of perceptrons actually got started in the 1950s with the first flush of AI, but didn't become a practical reality until after the turn of this millennium, really after the turn of this particular decade, 2010, when we started to see not only very large big data sets emerge and new approaches for managing it all, like Hadoop, come to the fore. But we've seen artificial neural nets get more sophisticated in terms of their capabilities, and a new approach for doing machine learning, artificial neural networks, with deeper layers of perceptrons, neurons, called deep learning has come to the fore. With deep learning, you have new algorithms like convolutional neural networks, recurrent neural networks, generative adversarial neural networks. These are different ways of surfacing up higher level abstractions in the data, for example for face recognition and object recognition, voice recognition and so forth. These all depend on this new state of the art for machine learning called deep learning. So what we have now in the year 2017 is we have quite a mania for all things AI, much of it is focused on deep learning, much of it is focused on tools that your average data scientist or your average developer increasingly can use and get very productive with and build these models and train and test them, and deploy them into working applications like going forward, things like autonomous vehicles would be impossible without this. >> Right, and we'll get some of that. But so you're saying that machine learning is essentially math that infers patterns from data. And math, it's new math, math that's been around for awhile or. >> Yeah, and inferring patterns from data has been done for a long time with software, and we have some established approaches that in many ways predate the current vogue for neural networks. We have support vector machines, and decision trees, and Bayesian logic. These are different ways of approaches statistical for inferring patterns, correlations in the data. They haven't gone away, they're a big part of the overall AI space, but it's a growing area that I've only skimmed the surface of. >> And they've been around for many many years, like SVM for example. Okay, now describe further, add some color to deep learning. You sort of painted a picture of this sort of deep layers of these machine learning algorithms and this network with some depth to it, but help us better understand the difference between machine learning and deep learning, and then ultimately AI. >> Yeah, well with machine learning generally, you know, inferring patterns from data that I said, artificial neural networks of which the deep learning networks are one subset. Artificial neural networks can be two or more layers of perceptrons or neurons, they have relationship to each other in terms of their activation according to various mathematical functions. So when you look at an artificial neural network, it basically does very complex math equations through a combination of what they call scalar functions, like multiplication and so forth, and then you have these non-linear functions, like cosine and so forth, tangent, all that kind of math playing together in these deep structures that are triggered by data, data input that's processed according to activation functions that set weights and reset the weights among all the various neural processing elements, that ultimately output something, the insight or the intelligence that you're looking for, like a yes or no, is this a face or not a face, that these incoming bits are presenting. Or it might present output in terms of categories. What category of face is this, a man, a woman, a child, or whatever. What I'm getting at is that so deep learning is more layers of these neural processing elements that are specialized to various functions to be able to abstract higher level phenomena from the data, it's not just, "Is this a face," but if it's a scene recognition deep learning network, it might recognize that this is a face that corresponds to a person named Dave who also happens to be the father in the particular family scene, and by the way this is a family scene that this deep learning network is able to ascertain. What I'm getting at is those are the higher level abstractions that deep learning algorithms of various sorts are built to identify in an automated way. >> Okay, and these in your view all fit under the umbrella of artificial intelligence, or is that sort of an uber field that we should be thinking of. >> Yeah, artificial intelligence as the broad envelope essentially refers to any number of approaches that help machines to think like humans, essentially. When you say, "Think like humans," what does that mean actually? To do predictions like humans, to look for anomalies or outliers like a human might, you know separate figure from ground for example in a scene, to identify the correlations or trends in a given scene. Like I said, to do categorization or classification based on what they're seeing in a given frame or what they're hearing in a given speech sample. So all these cognitive processes just skim the surface, or what AI is all about, automating to a great degree. When I say cognitive, but I'm also referring to affective like emotion detection, that's another set of processes that goes on in our heads or our hearts, that AI based on deep learning and so forth is able to do depending on different types of artificial neural networks are specialized particular functions, and they can only perform these functions if A, they've been built and optimized for those functions, and B, they have been trained with actual data from the phenomenon of interest. Training the algorithms with the actual data to determine how effective the algorithms are is the key linchpin of the process, 'cause without training the algorithms you don't know if the algorithm is effective for its intended purpose, so in Wikibon what we're doing is in the whole development process, DevOps cycle, for all things AI, training the models through a process called supervised learning is absolutely an essential component of ascertaining the quality of the network that you've built. >> So that's the calibration and the iteration to increase the accuracy, and like I say, the quality of the outcome. Okay, what are some of the practical applications that you're seeing for AI, and ML, and DL. >> Well, chat bots, you know voice recognition in general, Siri and Alexa, and so forth. Without machine learning, without deep learning to do speech recognition, those can't work, right? Pretty much in every field, now for example, IT service management tools of all sorts. When you have a large network that's logging data at the server level, at the application level and so forth, those data logs are too large and too complex and changing too fast for humans to be able to identify the patterns related to issues and faults and incidents. So AI, machine learning, deep learning is being used to fathom those anomalies and so forth in an automated fashion to be able to alert a human to take action, like an IT administrator, or to be able to trigger a response work flow, either human or automated. So AI within IT service management, hot hot topic, and we're seeing a lot of vendors incorporate that capability into their tools. Like I said, in the broad world we live in in terms of face recognition and Facebook, the fact is when I load a new picture of myself or my family or even with some friends or brothers in it, Facebook knows lickity-split whether it's my brother Tom or it's my wife or whoever, because of face recognition which obviously depends, well it's not obvious to everybody, depends on deep learning algorithms running inside Facebook's big data network, big data infrastructure. They're able to immediately know this. We see this all around us now, speech recognition, face recognition, and we just take it for granted that it's done, but it's done through the magic of AI. >> I want to get to the development angle scenario that you specialize in. Part of the reason why you came to Wikibon is to really focus on that whole application development angle. But before we get there, I want to follow the data for a bit 'cause you mentioned that was really the catalyst for the resurgence in AI, and last week at the Wikibon research meeting we talked about this three-tiered model. Edge, as edge piece, and then something in the middle which is this aggregation point for all this edge data, and then cloud which is where I guess all the deep modeling occurs, so sort of a three-tier model for the data flow. >> John: Yes. >> So I wonder if you could comment on that in the context of AI, it means more data, more I guess opportunities for machine learning and digital twins, and all this other cool stuff that's going on. But I'm really interested in how that is going to affect the application development and the programming model. John Farrier has a phrase that he says that, "Data is the new development kit." Well, if you got all this data that's distributed all over the place, that changes the application development model, at least you think it does. So I wonder if you could comment on that edge explosion, the data explosion as a result, and what it means for application development. >> Right, so more and more deep learning algorithms are being pushed to edge devices, by that I mean smartphones, and smart appliances like the ones that incorporate Alexa and so forth. And so what we're talking about is the algorithms themselves are being put into CPUs and FPGAs and ASICs and GPUs. All that stuff's getting embedded in everything that we're using, everything's that got autonomous, more and more devices have the ability if not to be autonomous in terms of making decisions, independent of us, or simply to serve as augmentation vehicles for our own whatever we happen to be doing thanks to the power of deep learning at the client. Okay, so when deep learning algorithms are embedded in say an internet of things edge device, what the deep learning algorithms are doing is A, they're ingesting the data through the sensors of that device, B, they're making inferences, deep learning algorithmic-driven inferences, based on that data. It might be speech recognition, face recognition, environmental sensing and being able to sense geospatially where you are and whether you're in a hospitable climate for whatever. And then the inferences might drive what we call actuation. Now in the autonomous vehicle scenario, the autonomous vehicle is equipped with all manner of sensors in terms of LiDAR and sonar and GPS and so forth, and it's taking readings all the time. It's doing inferences that either autonomously or in conjunction with inferences that are being made through deep learning and machine learning algorithms that are executing in those intermediary hubs like you described, or back in the cloud, or in a combination of all of that. But ultimately, the results of all those analytics, all those deep learning models, feed the what we call actuation of the car itself. Should it stop, should it put on the brakes 'cause it's about to hit a wall, should it turn right, should it turn left, should it slow down because it happened to have entered a new speed zone or whatever. All of the decisions, the actions that the edge device, like a car would be an edge device in this scenario, are being driven by evermore complex algorithms that are trained by data. Now, let's stay with the autonomous vehicle because that's an extreme case of a very powerful edge device. To train an autonomous vehicle you need of course lots and lots of data that's acquired from possibly a prototype that you, a Google or a Tesla, or whoever you might be, have deployed into the field or your customers are using, B, proving grounds like there's one out by my stomping ground out in Ann Arbor, a proving ground for the auto industry for self-driving vehicles and gaining enough real training data based on the operation of these vehicles in various simulated scenarios, and so forth. This data is used to build and iterate and refine the algorithms, the deep learning models that are doing the various operations of not only the vehicles in isolation but the vehicles operating as a fleet within an entire end to end transportation system. So what I'm getting at, is if you look at that three-tier model, then the edge device is the car, it's running under its own algorithms, the middle tier the hub might be a hub that's controlling a particular zone within a traffic system, like in my neck of the woods it might be a hub that's controlling congestion management among self-driving vehicles in eastern Fairfax County, Virginia. And then the cloud itself might be managing an entire fleet of vehicles, let's say you might have an entire fleet of vehicles under the control of say an Uber, or whatever is managing its own cars from a cloud-based center. So when you look at the tiering model that analytics, deep learning analytics is being performed, increasingly it will be for various, not just self-driving vehicles, through this tiered model, because the edge device needs to make decisions based on local data. The hub needs to make decisions based on a wider view of data across a wider range of edge entities. And then the cloud itself has responsibility or visibility for making deep learning driven determinations for some larger swath. And the cloud might be managing both the deep learning driven edge devices, as well as monitoring other related systems that self-driving network needs to coordinate with, like the government or whatever, or police. >> So envisioning that three-tier model then, how does the programming paradigm change and evolve as a result of that. >> Yeah, the programming paradigm is the modeling itself, the building and the training and the iterating the models generally will stay centralized, meaning to do all these functions, I mean to do modeling and training and iteration of these models, you need teams of data scientists and other developers who are both adept as to statistical modeling, who are adept at acquiring the training data, at labeling it, labeling is an important function there, and who are adept at basically developing and deploying one model after another in an iterative fashion through DevOps, through a standard release pipeline with version controls, and so forth built in, the governance built in. And that's really it needs to be a centralized function, and it's also very compute and data intensive, so you need storage resources, you need large clouds full of high performance computing, and so forth. Be able to handle these functions over and over. Now the edge devices themselves will feed in the data in just the data that is fed into the centralized platform where the training and the modeling is done. So what we're going to see is more and more centralized modeling and training with decentralized execution of the actual inferences that are driven by those models is the way it works in this distributive environment. >> It's the Holy Grail. All right, Jim, we're out of time but thanks very much for helping us unpack and giving us the skinny on machine learning. >> John: It's a fat stack. >> Great to have you in the office and to be continued. Thanks again. >> John: Sure. >> All right, thanks for watching everybody. This is Dave Vellante with Jim Kobelius, and you're watching theCUBE at the Marlboro offices. See ya next time. (upbeat music)

>> Narrator: From the Fairmont Hotel in the heart of Silicon Valley, it's theCUBE. Covering When IoT met AI: The Intelligence of Things. Brought to you by Western Digital. >> Hey welcome back here everybody Jeff Frick here with theCUBE. We're in downtown San Jose at the Fairmont Hotel. At a little event it's When IoT Met AI: The Intelligence of Things. As we hear about the Internet of Things all the time this is really about the data elements behind AI, and machine learning, and IoT. And we're going to get into it with some of the special guests here. We're excited to get the guy that's going to kick off this whole program shortly is Tom Stuermer. He is the I got to get the new title, the Global Managing Director, Ecosystem and Partnership, from Accenture. Tom, welcome-- >> Thank you, Jeff. >> And congrats on the promotion. >> Thank you. >> So IoT, AI, buzz words, a lot of stuff going on but we're really starting to see stuff begin to happen. I mean there's lots of little subtle ways that we're seeing AI work its way in to our lives, and machine learning work our way into its life, but obviously there's a much bigger wave that's about to crest here, shortly. So as you kind of look at the landscape from your point of view, you get to work with a lot of customers, you get to see this stuff implemented in industry, what's kind of your take on where we are? >> Well, I would say that we're actually very early. There are certain spaces with very well-defined parameters where AI's been implemented successfully, industrial controls on a micro level where there's a lot of well-known parameters that the systems need to operate in. And it's been very easy to be able to set those parameters up. There's been a lot of historical heuristic systems to kind of define how those work, and they're really replacing them with AI. So in the industrial spaces a lot of take up and we'll even talk a little bit later about Siemens who's really created a sort of a self-managed factory. Who's been able to take that out from a tool level, to a system level, to a factory level, to enable that to happen at those broader capabilities. I think that's one of the inflection points we're going to see in other areas where there's a lot more predictability and a lot of other IoT systems. To be able to take that kind of system level and larger scale factors of AI and enable prediction around that, like supply chains for example. So we're really not seeing a lot of that yet, but we're seeing some of the micro pieces being injected in where the danger of it going wrong is lower, because the training for those systems is very difficult. >> It's interesting, there's so much talk about the sensors, and the edge, and edge computing, and that's interesting. But as you said it's really much more of a system approach is what you need. And it's really kind of the economic boundaries of the logical system by which you're trying to make a decision in. We talk all the time, we optimizing for one wind turbine? Are you optimizing for one field that contains so many wind turbines? Are you optimizing for the entire plant? Or are you optimizing for a much bigger larger system that may or may not impact what you did on that original single turbine? So a systems approach is a really critical importance. >> It is and what we've seen is that IoT investments have trailed a lot of expectations as to when they were going to really jump in the enterprise. And what we're finding is that when we talk to our customers a lot of them are saying, look I've already got data. I've got some data. Let's say I'm a mining company and I've got equipment down in mines, I've got sensors around oxygen levels, I just don't get that much value from it. And part of the challenge is that they're looking at it from a historical data perspective. And they're saying well I can see the trajectory over time of what's happening inside of my mind. But I haven't really been able to put in prediction. I haven't been able to sort of assess when equipment might fail. And so we're seeing that when we're able to show them the ability to affect an eventual failure that might shut down revenue for a day or two when some significant equipment fails, we're able to get them to start making those investments and they're starting to see the value in those micro pockets. And so I think we're going to see it start to propagate itself through in a smaller scale, and prove itself, because there's a lot of uncertainty. There's a lot of work that's got to be done to stitch them together, and IoT infrastructure itself is already a pretty big investment as it is. >> Short that mine company, because we had Caterpillar on a couple weeks ago and you know their driving fleets of autonomous vehicles, they're talking about some of those giant mining trucks who any unscheduled downtime the economic impact is immense well beyond worrying about a driver being sick, or had a fight with his wife, or whatever reason is bringing down the productivity of those vehicles. So it's actually amazing the little pockets where people are doing it. I'm curious to get your point of view too on kind of you managed to comment the guy's like I'm not sure what the value is because the other kind of big topic that we see is when will the data and the intelligence around the data actually start to impact the balance sheet? Because data used to be kind of a pain, right? You had to store it, and keep it, and it cost money, and you had to provision servers, and storage, but really now and the future the data that you have, the algorithms you apply to it will probably be an increasing percentage of your asset value if not the primary part of you asset value, you seeing some people start to figure that out? >> Well they are. So if you look, if step back away from IoT for a minute and you look at how AI is being applied more broadly, we're finding some transformational value propositions that are delivering a lot of impacts to the bottom line. And it's anywhere from where people inside of a company interact with their customers, being able to anticipate their next move, being able to predict given these parameters of this customer what kind of customer care agent should I put on the phone with them before you even pick up the phone to anticipate some of those expectations. And we're seeing a lot of value in things like that. And so, excuse me, and so when you zoom it back in to IoT some of the challenges are that the infrastructure to implement IoT is very fragmented. There's 360 some IoT platform providers out in the world and the places where we're seeing a lot of traction in using predictive analytics and AI for IoT is really coming in the verticals like industrial equipment manufacturers where they've kind of owned the stack and they can define everything from the bottom up. And what they're actually being able to do is to start to sell product heavy equipment by the hour, by the use, because they're able to get telemeter off of that product, see what's happening, be able to see when a failure is about to come, and actually sell it as a service back to a customer and be able to predictably analyze when something fails and get spares there in time. And so those are some of the pockets where it's really far ahead because they've got a lot of vertical integration of what's happening. And I think the challenge on adoption of broader scale for companies that don't sell very expensive assets into the market is how do I as a company start to stitch my own assets that are for all kinds of different providers, and all kinds of the different companies, into a single platform? And what the focus has really been in IoT lately for the past couple of years is what infrastructure should I place to get the data? How do I provision equipment? How do I track it? How do I manage it? How do I get the data back? And I think that's necessary but completely insufficient to really get a lot of value IoT, because really all your able to do then is get data. What do you do with it? All the value is really in the data itself. And so the alternative approach a lot of companies are taking is starting to attack some of these smaller problems. And each one of them tends to have a lot of value on its own, and so they're really deploying that way. And some of them are looking for ways to let the battles of the platforms, let's at least get from 360 down to 200 so that I can make some bets. And it's actually proving to be a value, but I think that is one of the obstacles that we have to adoption. >> The other thing you mentioned interesting before we turned on the cameras is really thinking about AI as a way to adjust the way that we interact with the machines. There's two views of the machines taking over the world, is it the beautiful view, or we can freeze this up to do other things? Or certainly nobody has a job, right? The answer is probably somewhere in the middle. But clearly AI is going to change the way, and we're starting to see just the barely the beginnings with Alexa, and Siri, and Google Home, with voice interfacing and the way that we interact with these machines which is going to change dramatically with the power of, as you said, prescriptive analytics, presumptive activity, and just change that interaction from what's been a very rote, fixed, hard to change to putting as you said, some of these lighter weight, faster to move, more agile layers on the top stack which can still integrate with some of those core SAP systems, and systems of record in a completely different way. >> Exactly, you know I often use the metaphor of autonomous driving and people seem to think that that's kind of way far out there. But if you look at how driving an autonomous vehicle's so much different from driving a regular car, right? You have to worry about at the minutia of executing the driving process. You don't have to worry about throttle, break. You'd have to worry about taking a right turn on red. You'd have to worry about speeding. What you have to worry about is the more abstract concepts of source, destination, route that I might want to take. You can offload that as well. And so it changes what the person interacting with the AI system is actually able to do, and the level of cognitive capability that they're able to exercise. We're seeing similar things in medical treatment. We're using AI to do predictive analytics around injury coming off of medical equipment. It's not only starting to improve diagnoses in certain scenarios, but it's also enabling the techs and the doctors involved in the scans to think on a more abstract level about what the broader medical issues are. And so it's really changing sort of the dialogue that's happening around what's going on. And I think this is a good metaphor for us to look at when we talk about societal impacts of AI as well. Because there are some people who embrace moving forward to those higher cognitive activities and some who resist it. But I think if you look at it from a customer standpoint as well, no matter what business you're in if you're a services business, if you're a product business, the way you interact with your employees and the way you interact with your customers can fundamentally be changed with AI, because AI can enable the technology to bend it to your intentions. Someone at the call center that we talked about. I mean those are subtle activities. It's not just AI for voice recognition, but it's also using AI to alter what options are given to you, and what scenarios are going to be most beneficial. And more often than not you get it right. >> Well the other great thing about autonomous vehicles, it's just a fun topic because it's something that people can understand, and they can see, and they can touch in terms of a concept to talk about, some of these higher level concepts. But the second order impacts which most people don't even begin to think, they're like I want to drive my car is, you don't need parking lots anymore because the cars can all park off site. Just Like they do at airports today at the rental car agency. You don't need to build a crash cage anymore, because the things are not going to crash that often compared to human drivers. So how does the interior experience of a car change when you don't have to build basically a crash cage? I mean there's just so many second order impacts that people don't even really begin to think about. And we see this time and time again, we saw it with cloud innovation where it's not just is it cheaper to rent a server from Amazon than to buy one from somebody else? It's does the opportunity for innovation enable more of your people to make more contributions than they could before because they were too impatient to wait to order the server from the IT guy? So that's where I think too people so underestimate kind of the big Moore's Law my favorite, we overestimate in the short term and completely underestimate in the long term, the impacts of these things. >> It's the doubling function, exactly. >> Jeff: Yeah, absolutely. >> I mean it's hard for people, human kind is geared towards linear thinking, and so when something like Moore's Law continues to double every 18 months price performance continues to increase. Storage, compute, visualization, display. >> Networking, 5G. >> You know the sensors in MEMS, all of these things have gotten so much cheaper. It's hard for human of any intelligence to really comprehend what happens when that doubling occurs for the next 20 years. Which we're now getting on the tail end of that fact. And so those manifest themselves in ways that are a little bit unpredictable, and I think that's going to be one of our most exciting challenges over the next five years is what does an enterprise look like? What does a product look like? One of the lessons that, I spent a lot of time in race car engineering in my younger days and actually did quants and analytics, what we learned from that point is as you learned about the data you started to fundamentally change the architecture of the product. And I think that's going to be a whole new series of activities that are going to have to happen in the marketplace. Is people rethinking fundamental product. There's a great example of a company that's completely disrupted an industry. On the surface of it it's been disrupted because of the fact that they essentially disassociated the consumption from the provision of the product. And didn't have to own those assets so they could grow rapidly. But what they fundamentally did was to use AI to be able to broker when should I get more cars, where should the cars go? And because they're also we're on the forefront of being able to drive, this whole notion of consumption of cars, and getting people's conceptual mindset shifted to having owned a car to I know an Uber's going to be there. It becomes like a power outlet. I can just rely on it. And now people are actually starting to double think about should I even own a car? >> Whole different impact of the autonomous vehicles. And if I do own a car why should it be sitting in the driveway when I'm not driving it? Or I send it out to go work for me make it a performing asset. Well great conversation. You guys Accenture's in a great spot. You're always at the cutting edge. I used to tease a guy I used to work with at Accenture you've got to squeeze out all the fat in the supply chain (laughs) your RP days and again a lot of these things are people changing the lens and seeing fat and inefficiency and then attacking it in a different way whether it's Uber, Airbnb, with empty rooms in people's houses. We had Paul Doherty on at the GE Industrial Internet launch a few years back, so you guys are in a great position because you get to sit right at the forefront and help these people make those digital transformations. >> I appreciate that. >> I will tell you I mean supply chains is another one of those high level systems opportunities for AI where being able to optimize, think about it a completely automated distribution chain from factory all the way to the drone landing at your front doorstep as a consumer. That's a whole nother level of efficiency that we can't even contemplate right now. >> Don't bet against Bezos that's what I always say. All right, Tom Stuermer thanks for spending a few minutes and good luck with the keynote. >> I appreciate it Jeff. >> All right, I'm Jeff Frick you're watching theCUBE. We are at The Intelligence of Things, When IoT met AI. You're watching theCUBE. Thanks for watching. (upbeat music)

>> Male Voice: From the Fairmont Hotel in the heart of Silicon Valley, it's the Cube covering when IoT Met AI: the Intelligence of Things brought to you by Western Digital. >> Hey welcome back here everybody. Jeff Frick with the Cube. We're in downtown San Jose at the Fairmont Hotel at a small event talking about data and really in IoT and the intersection of all those things and we're excited to have a little startup boutique here and one of the startups is great enough to take the time to sit down with us. This is Shaun Moore, he's the founder and CEO of the recently renamed Trueface.ai. Shaun, welcome. >> Thank you for having me. >> So you've got a really cool company, Trueface,ai. I looked at the site. You have facial recognition software so that's cool but what I think is really more interesting is you're really doing facial recognition as a service. >> Shaun: Yes. >> And you a have a freemium model so I can go in and connect to your API and basically integrate your facial recognition software into whatever application that I built. >> Right so we were thinking about what we wanted to do in terms of pricing structure. We wanted to focus on the developer community so we wanted tinkers, people that just want to play with technology to help us improve it and then go after the kind of bigger clients and so we'll be hosting hack-a-thons. We just actually had one this past week in San Francisco. We had great feedback. We're really trying to get a base of you know, almost outsource engineers to help us improve this technology and so we have to offer it to them for free so we can see what they build from there. >> Right but you don't have an opensource component yet so you haven't gone that route? >> Not quite yet, no. >> Okay. >> We're thinking about that though. >> Okay, and still really young company, angel-funded, haven't taken it the institutional route yet. >> Right, yeah, we've been around since 2013, end of 2013, early 2014, and we were building smart home hardware so we had built the technology around originally to be a smart doorbell that used facial recognition to customize the smart home. From the the trajectory went, we realized our clients were using it more for security purposes and access-control, not necessarily personalization. We made a quick pivot to a quick access control company and continue to learn about how people are using facial recognition in practice. Could it be a commercial technology that people are comfortable with? And throughout that thought process and going through and testing a bunch of other facial recognition technologies, we realized we could actually build our own platform and reach a larger audience with it and essentially be the core technology of a lot cooler and more innovative products. >> Right, and not get into the hardware business of doorbells >> Yeah, the hardware business is tough. >> That's a tough one. >> We were going to through manufacturing one and I'm glad we don't have to do that again. >> So what are some of the cool ways that people are using facial recognition that maybe we would never have thought about? >> Sure, so for face matching - The API is four components. It's face matching, face detection, face identification, and what we call spoof detection. Face matching is what it sounds like: one-to-one matching. Face detection is just detecting that someone is in the frame. The face identification is your one to act so your going into a database of people. And your spoof detection is if someone holds up a picture of me or of you and tries to get it, we'll identify that as an attack attempt and that's kind of where we differentiate our technology from most is not a lot of technology out there can do that piece and so we've packaged that all up into essentially the API for all these developers to use and some of the different ideas that people have come up with for us have been for banking logins, so for ATMs, you walk up to an ATM, you put your card in and set up a PIN so to prevent against fraud it actually scans your face and does a one-to-one match. For ship industries, so for things like cruise ships, when people get off and then come back on, instead of having them show ID, they use quick facial recognition scans. So we're seeing a lot of different ideas. One of the more funny ones is based off a company out in LA that is doing probation monitoring for drunk drivers and so we've built technology that's drunk or not drunk. >> Drunk or not drunk? >> Right so we can actually measure based on historical data if your face appears to be drunk and so you know, the possibilities are truly endless. And that's why I said we went after the development community first because >> Right right >> They're coming to use with these creative ideas. >> So it's interesting with this drunk or not drunk, of course, not to make fun of drunk driving, it's not a funny subject but obviously you've got an algorithm that determines anchor points on the eyes and the nose and certain biometric features but drunk, you're looking for much softer, more subtle clues, I would imagine because the fundamental structure of your face hasn't changed. >> Right so it's a lot of training data, so it's a lot of training data. >> Well a lot of training data, yeah. We don't want to go down that path. >> So a lot of research on our team's part. >> Well then the other thing too is the picture, is the fraud attempt. You must be looking around and shadowing and really more 3D-types of things to look over something as simple as holding up a 2D picture. >> Right so a lot of the technology that's tried to do it, that's tried to prevent against picture attacks has done so with extra hardware or extra sensors. We're actually all cloud-based right now so it isn't our software and that is what is special to us is that picture attack detection but we've a got a very very intelligent way to do it. Everything is powered by deep learning so we're constantly understanding the surroundings, the context, and making an analysis on that. >> So I'm curious from the data side, obviously you're pulling in kind of your anchor data and then for doing comparisons but then are you constantly updating that data? I mean, what's kind of your data flow look like in terms of your algorithms, are you constantly training them and adjusting those algorithms? How does that work kind of based on real time data versus your historical data? >> So we have to continue to innovate and that is how we do it, is we continue to train every single time someone shows up we train their profile once more and so if you decide to grow a beard, you're not going to grow a beard in one day, right? It's going to take you a week, two weeks. We're learning throughout those two weeks and so it's just a way for use to continue to get more data for us but also to ensure that we are identifying you properly. >> Right, do you use any external databases that you pull in as some type of you know, adding more detail or you know, kind of, other public sources or it's all your own? >> It's all our own. >> Okay and I'm curious too on the kind of opening up to the developer community, how has that kind of shaped your product roadmap and your product development? >> It - we've got to be very very conscious of not getting sidetracked because we get to hear cool ideas about what we could do but we've got our core focus of building this API for more people to use. So you know, we continue to reach out them and ask for help and you know if they find flaw or they find something cool that we want to continue to improve, we'll keep working on that so I think it's more of a - we're finding the developer community likes to really tinker and to play and because they're doing it out of passion, it helps us drive our product. >> Right right. Okay, so priorities for the rest of the year? What's at the top of the list? >> We'll be doing a bigger rollout with a couple of partners later on this year and those will be kind of our flagship partners. But again, like I said, we want to continue to support those development communities so we'll be hosting a lot of hack-a-thons and just really pushing the name out there. So we launched our product yesterday and that helped generate some awareness but we're going to have to continue to have to get the brand out there as it's now one day old. >> Right right, well good. Well it was Chui before and it's Trueface.ai so we look forward to keeping an eye on progress and congratulations on where you've gotten to date. >> Thank you very much. I appreciate that. >> Absolutely. Alrighty, Shaun Moore, it's Trueface.ai. Look at the cameras, smile, it will know it's you. You're watching Jeff Frick down at the Cube in downtown San Jose at the When IoT Met AI: The Intelligence of Things. Thanks for watching. We'll be right back after this short break.

>> Announcer: From the Fairmont Hotel in the heart of Silicon Valley, it's The Cube. Covering when IOT met AI, the intelligence of things. Brought to you by Western Digital. >> Hey, welcome back everybody. Jeff Rick here with The Cube. We're in downtown San Jose at the Fairmont Hotel at a little show called when IOT Met AI, the Intelligence of Things. Talking about big data, IOT, AI and how those things are all coming together with virtual reality, artificial intelligence, augmented reality, all the fun buzz words, but this is where it's actually happening and we're real excited to have a pioneer in this space. He's Jack McCauley. He was a co-founder at Occulus VR, now spending his time at UC Berkeley as an innovator in residence. Jack welcome. >> Thank you. >> So you've been watching this thing evolve, obviously Occulus, way out front in kind of the VR space and I think augmented a reality in some ways is even more exciting than just kind of pure virtual reality. >> Right. >> So what do you think as you see this thing develop from the early days when you first sat down and started putting this all together? >> Well, I come from a gaming background. That's what I did for 30 years. I worked in video game development, particularly in hardware and things, console hardware. >> That's right, you did the Guitar Hero. >> Guitar Hero. Yeah, that's right. >> We got that one at home. >> I built their guitars and designed and built their guitars for Activision. And when were part of Red Octane, which is a studio. I primarily worked in the studio, not the headquarters, but I did some of the IP work with them too, so, to your question, you know when you produce a product and put it on the market, you never really know how it's going to do. >> Jeff: Right. >> So we make, we made two developer kits, put them out there and they exceeded our expectations and that was very good. It means that there is a market for VR, there is. We produce a consumer version and sales are not what we expected for that particular product. That was designated towards PC gamers and hopefully console games. But what has done well is the mobile stuff has exceeded everyone's mildest expectations. I heard numbers, Gear VR, which is Occulus designed product for me, sold 7 million of those. That's a smash hit. Now, worldwide for phone mounted VR goggles, it's about 20 million and that's just in two years, so that's really intriguing. So, what has happened is it's shifted away from an expensive PC based rig with $700 or whatever it costs, plus $1,500 for the computer to something that costs $50 and you just stick your cell phone in it and that's what people, it doesn't give you the best experience, but that's what has sold and so if I were doing a start-up right now, I would not be working on PC stuff, I'd be working on mobile stuff. >> Jeff: Right. >> And the next thing I think, which will play out of this is, and I think you mentioned it prior to the interview, is the 360 cameras and Google has announced a camera that they're going to come out and it's for their VR 180 initiative, which allows you to see 180 video in stereo with a cell phone strapped to your face. And that's very intriguing. There's a couple of companies out there working on similar products. Lucid Cam, which is a start-up company here has a 180 camera that's very, very good and they have one coming out that's in 4K. They just launched their product. So to answer your question, it looks like what is going to happen is for VR, is that it's a cell phone strapped to your face and a camera somewhere else that you can view and experience. A concert. Imagine taking it to a sporting event where 5,000 people can view your video, 10,000 from your seat. That's very intriguing. >> Yeah, it's interesting I had my first kind of experience just not even 360 or live view, but I did a periscope from the YouTube concert here at Levi Stadium a couple of months ago, just to try it out, I'd never really done it and it was fascinating to watch the engagement of people on that application who had either seen them the prior week in Seattle or were anticipating them coming to the Rose Bowl, I think, you know, within a couple of days, and to have an interaction just based on my little, you know, mobile phone, I was able to find a rail so I had a pretty steady vantage point, but it was a fascinating, different way to experience media, as well as engagement, as well as kind of a crowd interaction beyond the people that happened to be kind of standing in a circle. >> You, what's intriguing about VR 180 is that anybody can film the concert and put the video on YouTube or stream it through their phone. And formerly it would require a $10,000 camera, a stereo camera set up professionally, but can you imagine though that a crowd, you know, sourced sort of thing where the media is sourced by the crowd and anyone can watch it with a mobile phone. That's what's happening, I think, and with Google's announcement, it even that reinforces my opinion anyways that that is where the market will be. It's live events, sporting events. >> Right, it's an experience, right? It all comes back to kind of experience. People are so much more experience drive these days than I think thing driven from everything from buying cars versus taking a new Uber and seeing it over and over and over again. People want the experience, but not necessarily, as the CEO of Zura said, the straps and straddles of ownership, let me have the fun, I don't necessarily want to own it. But I think the other thing that gets less talked about, get your opinion, is really the kind of combination of virtual reality plus the real world, augmented reality. We see the industrial internet of things all the time where, you know, you go take a walk on that factory before you put your goggles on and not only do you see what you see that's actually in front of you, but now you can start to see, it's almost like a heads up display, certain characteristics of the machinery and this and that are now driven from the database side back into the goggles, but now the richness of your observation has completely changed. >> Yes, and in some ways when you think of what Google did with Google Glass, not as well as we had liked. >> But for a first attempt. >> Yeah. They're way ahead of their time and there will come a time when, you know, Snap has their specs, right? Have you seen those? It's not augmented reality, but, there will come a time when you can probably have a manacle on your face and see the kinds of things you need to see if your driving a car for instance that, I mean, a heads up display or a projector projecting right into your retina. So, and, so I think that's the main thing for augmented reality. Will people, I mean, your Pokemon Go, that's kind of a AR game in a way. You look through your cell phone and the character stays fixed on the table or wherever you're looking for it. I mean that uses a mobile device to do that and I can imagine other applications that use a mobile device to do that and I'm aware of people working on things like that right now. >> So do you think that the breakthrough on the mobile versus the PC-based system was just good enough? In being able to just experience that so easily, you know, I mean, Google gave out hundreds and hundreds of thousands of the cardboard boxes, so wow. >> Yeah. Well, it didn't mean that Gear VR didn't move into the market, it did. You know, it did anyways, but to answer your question about AR, you know, I think that, you know, without having good locals, I mean the problem with wearing the Google Glass and the Google cardboard and Gear VR is it kind of makes you sick a little bit and nobody's working on the localization part. Like how to get rid of the nausea effect. I watched a video that was filmed with Lucid Cam at the Pride Parade in San Francisco and I put it on and somebody was moving with the crowd and I just felt nauseous, so that problem probably probably is one I would attempt to attack if I were going to build a company or something like that right now. >> But I wonder too, how much of that is kind of getting used to the format because people when they first put them on for sure, there's like, ah, but you know, if you settle in a little bit and our eyes are pretty forgiving, you get used to things pretty quickly. Your mind can get accustomed to it to a certain degree, but even I get nauseous and I don't get nauseous very easily. >> Okay, so you're title should just be tinkerer. I looked at your Twitter handle. You're building all kinds of fun stuff in your not a garage, but your big giant lab and you're working at Berkeley. What are some of the things that you can share that you see coming down the road that people aren't necessary thinking about that's going to take some of these technologies to the next level. >> I got one for you. So you've heard of autonomous vehicles, right? >> Jeff: Yep, yep. >> And you've heard of Hollow Lens, right. Hollow Lens is an augmented reality device you put on your had and it's got built in localization and it creates what's, it's uses what's know as SLAM or S-L-A-M to build a mesh of the world around you. And with that mesh, the next guy that comes into that virtual world that you mapped will be away ahead. In other words, the map will already exists and he'll modify upon that and the mesh always gets updated. Can you imagine getting that into a self-driving vehicle just for safety's sake, mapping out the road ahead of you, the vehicle ahead of you has already mapped the road for you and you're adding to the mesh and adjusting the mesh, so I think that that's, you know, as far as Hollow Lens is concerned and their localization system, that's going to be really relevant to self-driving cars. Now whether or not it'll be Microsoft's SLAM or somebody else's, I think that that's probably the best, that's the good thing that came out of Hollow Lens and that will bleed into the self-driving car market. It's a big data crunching number and in Jobs, he was actually looking at this a long time ago, like what can we do with self-driving vehicles and I think he had banned the idea because he realized he had a huge computing and data problem. That was 10 years ago. Things have changed. But I think that that's the thing that will possibly come out of, you know, this AR stuff is that localization is just going to be transported to other areas of technology and self-driving cars and so forth. >> I just love autonomous vehicles because everything gets distilled and applied into that application, which is a great application for people to see and understand it's so tangible. >> Yeah, it may change the way we think about cars and we may just not ever own a car. >> I think absolutely. The car industry, it's ownership, it's usage, it's frequency of usage, how they're used. It's not a steel cage anymore for safety as the crash rates go down significantly. I think there's a lot of changes. >> Yeah, you buy a car and it sits for 20 hours a day. >> Right. >> Unutilized. >> All right. Well, Jack I hope maybe I get a chance to come out and check out your lab one time because you're making all kinds of cool stuff. When's that car going to be done? >> I took it upon myself to remodel a house the same time I was doing that, but the car is moving ahead. In September I think I can get it started. Get the engine running and get the power train up and running. Right now I'm working on the electronics and we have an interesting feature on that car that we're going to do an announcement on later. >> Okay, we'll look out for that. We'll keep watching the Twitter. All right, thanks for taking a few minutes. All right, let's check with Cauley. I'm Jeff Rick. You're watching The Cube from When IOT Met AI, the Intelligence of Things in San Jose. We'll be right back after this short break. Thanks for watching. (technological jingle)

>> Presenter: From the Fairmont Hotel, in the heart of Silicon Valley, it's theCUBE. Covering When IoT Met AI The Intelligence of Things. Brought to you by Western Digital. >> Hey welcome back everybody, Jeff Frick here with theCUBE. We're in downtown San Jose at the Fairmont Hotel, at an event called When IoT Met AI The Intelligence of Things. You've heard about the internet of things, and on the intelligence of things, it's IoT, it's AI, it's AR, all this stuff is really coming to play, it's very interesting space, still a lot of start-up activity, still a lot of big companies making plays in this space. So we're excited to be here, and really joined by our host, big thanks to Western Digital for hosting this event with WDLabs' Dave Tang. Got newly promoted since last we spoke. The SVP of corporate marketing and communications, for Western Digital, Dave great to see you as usual. >> Well, great to be here, thanks. >> So I don't think the need for more storage is going down anytime soon, that's kind of my takeaway. >> No, no, yeah. If this wall of data just keeps growing. >> Yeah, I think the term we had yesterday at the Ag event that we were at, also sponsored by you, is really the flood of data using an agricultural term. But it's pretty fascinating, as more, and more, and more data is not only coming off the sensors, but coming off the people, and used in so many more ways. >> That's right, yeah we see it as a virtual cycle, you create more data, you find more uses for that data to harness the power and unleash the promise of that data, and then you create even more data. So, when that virtual cycle of creating more, and finding more uses of it, and yeah one of the things that we find interesting, that's related to this event with IoT and AI, is this notion that data is falling into two general categories. There's big data, and there's fast data. So, big data I think everyone is quite familiar with by this time, these large aggregated likes of data that you can extract information out of. Look for insights and connections between data, predict the future, and create more prescriptive recommendations, right? >> Right. >> And through all of that you can gain algorithms that help to make predictions, or can help machines run based on that data. So we've gone through this phase where we focused a lot on how we harness big data, but now we're taking these algorithms that we've gleaned from that, and we're able to put them in real time applications, and that's sort of been the birth of fast data, it's been really-- >> Right, the streaming data. We cover Spark Summit, we cover Flink, and New, a new kind of open source project that came out of Berlin. That some people would say the next generation of Spark, and the other thing, you know, good for you guys, is that it used to be, not only was it old data, but it was a sampling of old data. Now on this new data, and the data stream that's all of the data. And I would actually challenge, I wonder if that separation as you describe, will stay, because I got to tell you, the last little drive I bought, just last week, was an SSD drive, you know, one terabyte. I needed some storage, and I had a choice between spinning disc and not, and I went with the flat. I mean, 'cause what's fascinating to me, is the second order benefits that we keep hearing time, and time, and time again, once people become a data-driven enterprise, are way more than just that kind of top-level thing that they thought. >> Exactly, and that's sort of that virtual cycle, you got to taste, and you learn how to use it, and then you want more. >> Jeff: Right, right. >> And that's the great thing about the breadth of technologies and products that Western Digital has, is from the solid state products, the higher performance flash products that we have, to the higher capacity helium-filled drive technologies, as well as devices going on up into systems, we cover this whole spectrum of fast data and big data. >> Right, right. >> I'll give an example. So credit card fraud detection is an interesting area. Billions of dollars potentially being lost there. Well to learn how to predict when transactions are fraudulent, you have to study massive amounts of data. Billions of transactions, so that's the big data side of it, and then as soon as you do that, you can take those algorithms and run them in real time. So as transactions come in for authorization, those algorithms can determine, before they're approved, that one's fraudulent, and that one's not. Save a lot of time and processing for fraud claims. So that's a great example of once you learn something from big data, you apply it to the real-time realm, and it's quite dire right? And then that spawned you to collect even more data, because you want to find new applications and new uses. >> Right, and too kind of this wave of computing back and forth from the shared services computer, then the desktop computer, now it's back to the cloud, and then now it's-- >> Dave: Out with the edge. >> IoT, it's all about the edge. >> Yeah, right. >> And at the end of the day, it's going to be application-specific. What needs to be processed locally, what needs to be processed back at the computer, and then all the different platforms. We were again at a navigation for autonomous vehicles show, who knew there was such a thing that small? And even the attributes of the storage required in the ecosystem of a car, right? And the environmental conditions-- >> That's right. >> Is the word I'm looking for. Completely different, new opportunity, kind of new class of hardware required to operate in that environment, and again that still combines cloud and Edge, sensors and maps. So just the, I don't think that the man's going down David. >> Yeah, absolutely >> I think you're in a good spot. (Jeff laughing) >> You're absolutely right, and even though we try to simplify into fast data, and big data, and Core and Edge, what we're finding is that applications are increasingly specialized, and have specialized needs in terms of the type of data. Is it large amounts of data, is it streaming? You know, what are the performance characteristics, and how is it being transformed, what's the compute aspect of it? And what we're finding, is that the days of general-purpose compute and storage, and memory platforms, are fading, and we're getting into environments with increasingly specialized architectures, across all those elements. Compute, memory and storage. So that's what's really exciting to be in our spot in the industry, is that we're looking at creating the future by developing new technologies that continue to fuel that growth even further, and fuel the uses of data even further. >> And fascinating just the ongoing case of Moore's law, which I know is not, you know you're not making microprocessors, but I think it's so powerful. Moore's law really is a philosophy, as opposed to an architectural spec. Just this relentless pace of innovation, and you guys just continue to push the envelope. So what are your kind of priorities? I can't believe we're halfway through 2017 already, but for kind of the balance of the year kind of, what are some of your top-of-mind things? I know it's exciting times, you're going through the merger, you know, the company is in a great space. What are your kind of top priorities for the next several months? >> Well, so, I think as a company that has gone through serial acquisitions and integrations, of course we're continuing to drive the transformation of the overall business. >> But the fun stuff right? It's not to increase your staff (Jeff laughing). >> Right, yeah, that is the hardware. >> Stitching together the European systems. >> But yeah, the fun stuff includes pushing the limits even further with solid state technologies, with our 3D NAND technologies. You know, we're leading the industry in 64 layer 3D NAND, and just yesterday we announced a 96 layer 3D NAND. So pushing those limits even further, so that we can provide higher capacities in smaller footprints, lower power, in mobile devices and out on the Edge, to drive all these exciting opportunities in IoT an AI. >> It's crazy, it's crazy. >> Yeah it is, yeah. >> You know, terabyte SD cards, terabyte Micro SD cards, I mean the amount of power that you guys pack into these smaller and smaller packages, it's magical. I mean it's absolutely magic. >> Yeah, and the same goes on the other end of the spectrum, with high-capacity devices. Our helium-filled drives are getting higher and higher capacity, 10, 12, 14 terabyte high-capacity devices for that big data core, that all the data has to end up with at some point. So we're trying to keep a balance of pushing the limits on both ends. >> Alright, well Dave, thanks for taking a few minutes out of your busy day, and congratulations on all your success. >> Great, good to be here. >> Alright, he's Dave Tang from Western Digital, he's changing your world, my world, and everyone else's. We're here in San Jose, you're watching theCUBE, thanks for watching.

(upbeat electronic music) >> Narrator: From the Fairmont Hotel in the heart of Silicon Valley, it's theCUBE. Covering when IoT met AI, The Intelligence of Things. Brought to you by Western Digital. >> Welcome back here everybody, Jeff Frick here with theCUBE. We are at downtown San Jose at the Fairmont Hotel. When IoT met AI it happened right here, you saw it first. The Intelligence of Things, a really interesting event put on by readwrite and Western Digital and we are really excited to welcome back a many time CUBE alumni and always a fan favorite, she's Janet George. She's Fellow & Chief Data Officer of Western Digital. Janet, great to see you. >> Thank you, thank you. >> So, as I asked you when you sat down, you're always working on cool things. You're always kind of at the cutting edge. So, what have you been playing with lately? >> Lately I have been working on neural networks and TensorFlow. So really trying to study and understand the behaviors and patterns of neural networks, how they work and then unleashing our data at it. So trying to figure out how it's training through our data, how many nets there are, and then trying to figure out what results it's coming with. What are the predictions? Looking at how the predictions are, whether the predictions are accurate or less accurate and then validating the predictions to make it more accurate, and so on and so forth. >> So it's interesting. It's a different tool, so you're learning the tool itself. >> Yes. >> And you're learning the underlying technology behind the tool. >> Yes. >> And then testing it actually against some of the other tools that you guys have, I mean obviously you guys have been doing- >> That's right. >> Mean time between failure analysis for a long long time. >> That's right, that's right. >> So, first off, kind of experience with the tool, how is it different? >> So with machine learning, fundamentally we have to go into feature extraction. So you have to figure out all the features and then you use the features for predictions. With neural networks you can throw all the raw data at it. It's in fact data-agnostic. So you don't have to spend enormous amounts of time trying to detect the features. Like for example, If you throw hundreds of cat images at the neural network, the neural network will figure out image features of the cat; the nose, the eyes, the ears and so on and so forth. And once it trains itself through a series of iterations, you can throw a lot of deranged cats at the neural network and it's still going to figure out what the features of a real cat is. >> Right. >> And it will predict the cat correctly. >> Right. So then, how does that apply to, you know, the more specific use case in terms of your failure analysis? >> Yeah. So we have failures and we have multiple failures. Some failures through through the human eye, it's very obvious, right? But humans get tired, and over a period of time we can't endure looking at hundreds and millions of failures, right? And some failures are interconnected. So there is a relationship between these failure patterns or there is a correlation between two failures, right? It could be an edge failure. It could a radial failure, eye pattern type failure. It could be a radial failure. So these failures, for us as humans, we can't escape. >> Right. >> And we used to be able to take these failures and train them at scale and then predict. Now with neural networks, we don't have to take and do all that. We don't have to extract these labels and try to show them what these failures look like. Training is almost like throwing a lot of data at the neural networks. >> So it almost sounds like kind of the promise of the data lake if you will. >> Yes. >> If you have heard about, from the Hadoop Summit- >> Yes, yes, yes. >> For ever and ever and ever. Right? You dump it all in and insights will flow. But we found, often, that that's not true. You need hypothesis. >> Yes, yes. >> You need to structure and get it going. But what you're describing though, sounds much more along kind of that vision. >> Yes, very much so. Now, the only caveat is you need some labels, right? If there is no label on the failure data, it's very difficult for the neural networks to figure out what the failure is. >> Jeff: Right. >> So you have to give it some labels to understand what patterns it should learn. >> Right. >> Right, and that is where the domain experts come in. So we train it with labeled data. So if you are training with a cat, you know the features of a cat, right? In the industrial world, cat is really what's in the heads of people. The domain knowledge is not so authoritative. Like the sky or the animals or the cat. >> Jeff: Right. >> The domain knowledge is much more embedded in the brains of the people who are working. And so we have to extract that domain knowledge into labels. And then you're able to scale the domain. >> Jeff: Right. >> Through the neural network. >> So okay so then how does it then compare with the other tools that you've used in the past? In terms of, obviously the process is very different, but in terms of just pure performance? What are you finding? >> So we are finding very good performance and actually we are finding very good accuracy. Right? So once it's trained, and it's doing very well on the failure patterns, it's getting it right 90% of the time, right? >> Really? >> Yes, but in a machine learning program, what happens is sometimes the model is over-fitted or it's under-fitted or there is bias in the model and you got to remove the bias in the model or you got to figure out, well, is the model false-positive or false-negative? You got to optimize for something, right? >> Right, right. >> Because we are really dealing with mathematical approximation, we are not dealing with preciseness, we are not dealing with exactness. >> Right, right. >> In neural networks, actually, it's pretty good, because it's actually always dealing with accuracy. It's not dealing with precision, right? So it's accurate most of the time. >> Interesting, because that's often what's common about the kind of difference between computer science and statistics, right? >> Yes. >> Computers is binary. Statistics always has a kind of a confidence interval. But what you're describing, it sounds like the confidence is tightening up to such a degree that it's almost reaching binary. >> Yeah, yeah, exactly. And see, brute force is good when your traditional computing programing paradigm is very brute force type paradigm, right? The traditional paradigm is very good when the problems are simpler. But when the problems are of scale, like you're talking 70 petabytes of data or you're talking 70 billion roles, right? Find all these patterns in that, right? >> Jeff: Right. >> I mean you just, the scale at which that operates and at the scale at which traditional machine learning even works is quite different from how neural networks work. >> Jeff: Okay. >> Right? Traditional machine learning you still have to do some feature extraction. You still have to say "Oh I can't." Otherwise you are going to have dimensionality issues, right? It's too broad to get the prediction anywhere close. >> Right. >> Right? And so you want to reduce the dimensionality to get a better prediction. But here you don't have to worry about dimensionality. You just have to make sure the labels are right. >> Right, right. So as you dig deeper into this tool and expose all these new capabilities, what do you look forward to? What can you do that you couldn't do before? >> It's interesting because it's grossly underestimating the human brain, right? The human brain is supremely powerful in all aspects, right? And there is a great deal of difficulty in trying to code the human brain, right? But with neural networks and because of the various propagation layers and the ability to move through these networks we are coming closer and closer, right? So one example: When you think about driving, recently, Google driverless car got into an accident, right? And where it got into an accident was the driverless car was merging into a lane and there was a bus and it collided with the bus. So where did A.I. go wrong? Now if you train an A.I., birds can fly, and then you say penguin is a bird, it is going to assume penguin can fly. >> Jeff: Right, right. >> We as humans know penguin is a bird but it can't fly like other birds, right? >> Jeff: Right. >> It's that anomaly thing, right? Naturally when are driving and a bus shows up, even if it's yield, the bus goes. >> Jeff: Right, right. >> We yield to the bus because it's bigger and we know that. >> A.I. doesn't know that. It was taught that yield is yield. >> Right, right. >> So it collided with the bus. But the beauty is now large fleets of cars can learn very quickly based on what it just got from that one car. >> Right, right. >> So now there are pros and cons. So think about you driving down Highway 85 and there is a collision, it's Sunday morning, you don't know about the collision. You're coming down on the hill, right? Blind corner and boom that's how these crashes happen and so many people died, right? If you were driving a driverless car, you would have knowledge from the fleet and from everywhere else. >> Right. >> So you know ahead of time. We don't talk to each other when we are in cars. We don't have universal knowledge, right? >> Car-to-car communication. >> Car-to-car communications and A.I. has that so directly it can save accidents. It can save people from dying, right? But people still feel, it's a psychology thing, people still feel very unsafe in a driverless car, right? So we have to get over- >> Well they will get over that. They feel plenty safe in a driverless airplane, right? >> That's right. Or in a driveless light rail. >> Jeff: Right. >> Or, you know, when somebody else is driving they're fine with the driver who's driving. You just sit in the driver's car. >> But there's that one pesky autonomous car problem, when the pedestrian won't go. >> Yeah. >> And the car is stopped it's like a friendly battle-lock. >> That's right, that's right. >> Well good stuff Janet and always great to see you. I'm sure we will see you very shortly 'cause you are at all the great big data conferences. >> Thank you. >> Thanks for taking a few minutes out of your day. >> Thank you. >> Alright she is Janet George, she is the smartest lady at Western Digital, perhaps in Silicon Valley. We're not sure but we feel pretty confident. I am Jeff Frick and you're watching theCUBE from When IoT meets AI: The Intelligence of Things. We will be right back after this short break. Thanks for watching. (upbeat electronic music)

>> Announcer: From the Fairmont Hotel in the heart of Silicon Valley, it's The Cube. Covering When IoT met AI: The Intelligence of Things. Brought to you by Western Digital. >> Hey, welcome back, everybody. Jeff Frick here with The Cube. We're in downtown Los Angeles at the Fairmont Hotel at a interesting little show called When IoT Met AI: The Intelligence of Things. A lot of cool startups here along with some big companies. We're really excited go have our next guest, taking a little different angle. He's Scott Noteboom. He is the co-founder and CEO of a company called Litbit. First off, Scott, welcome. >> Yeah, thank you very much. >> Absolutely. For folks that aren't familiar, what is Litbit, what's your core mission? >> Well, probably, the simplest way to put it is, is in business we enable our users who have a lot of experience in a lot of different areas to take their expertise and experience which may not be coding software, or understanding, or even being able to spell what an algorithm is on the data science perspective, and being able to give them an easy interface so they can kind of create their own Siro or Alexa, an AI but an AI that's based on their own subject matter expertise that they can put to work in a lot of different ways. >> So, there's often a lot of talk about kind of tribal knowledge, and how does tribal knowledge get passed down so people know how to do things. Whether it's with new employees, or as you were talking about a little bit off camera, just remote locations for this or that. And there hasn't really been a great system to do that. So, you're really attacking that, not only with the documentation, but then making an AI actionable piece of software that can then drive machines and using IoT to do things. Is that correct? >> That's right. So, if you created, say an AI that I've been passionate about 'cause I ran data centers for a lot of years, is DAC. So, DAC's an AI that has a lot of expertise, and how to run a data center by, and kind of fueled and mentored by a lot of the experts in the industry. So, how can you take DAC and put Dak to work in a lot of places? And the people who need the best trained DAC aren't people who are building apps. They are people who have their area of subject matter expertise, and we view these AI personas that can be put to work as kind of apps of the future, where can people can prescribe to personas that are build directly by the experts, which is a pretty pure way to connect AIs with the right people, and then be able to get them and put them-- >> So, there's kind of two steps to the process. How does the information get from the experts into your system? How's that training happen? >> So, where we spend a lot of attention is, a lot of people question and go, "Well, an AI lives in this virtual logical world "that's disconnected from the physical world." And I always questions for people to close their eyes and imagine their favorite person that loves them in the world. And when they picture that person hear that person's voice in their head, that's actually a very similar virtual world as what AIs working. It's not the physical world. And what connects us as people to the physical world, our senses, our sight, our hearing, our touch, our feeling. And what we've done is we've enabled using IoT sensors, the ability of combining those sensors with AI to turn sensors into senses, which then provide the ability for the AI to connect really meaningful ways to the physical world. And then the experts can teach the AI this is what this looks like, this is what this sounds like, this is what it's supposed to feel like. If it's greater than 80 degrees in an office location, it's hot. Really teaching the AI to be able to form thoughts based on a specific expertise and then be able to take the right actions to do the right things when those thoughts are formed. >> How do you deal with nuance, 'cause I'm sure there's a lot of times where people, as you said, are sensing or smelling or something, but they don't even necessarily consciously know that that's an input into their decision process, even though it really is. They just haven't really thought of it as a discrete input. How do you separate out all these discreet inputs so you get a great model that represents your best of breed technicians? >> Well, to try to answer the question, first of all, the more training the better. So, the good way to think of the AI is, unlike a lot of technologies that typically age and go out of life over time, an AI continuously gets smarter the more it's mentored by people, which would be supervised learning. And the more it can adjust and learn on it's own combined with real day to day data activity combined with that supervised learning and unsupervised learning approach, so enabling it to continuously get better over time. We've figure out some ways that it can produce some pretty meaningful results with a small amount of training. So, yeah. >> Okay. What are some of the applications, kind of your initial go to market? >> We're a small startup, and really, what we've done is we've developed a platform that we really like to, our goal is for it to be very horizontal in nature. And then the applications or the AI personas can be very vertical or subject matter experts across different silos. So, what we're doing is, is we're working with partners right now in different silos developing AIs that have expertise in the oil and gas business, in the pharmaceutical space, in the data center space, in the corporate facilities manage space, and really making sure that people who aren't technologists in all of those spaces, whether you're a very specific scientists who're running a lab, or a facilities guy in a corporate building, can successfully make that experiential connection between themselves and the AI, and put it to practical use. And then as we go, there's a lot of efforts that can be very specific to specific silos, whatever they may be. >> So, those personas are actually roles of individuals, if you will, performing certain tasks within those verticals. >> Absolutely. What we call them is coworkers, and the way things are designed is, one of the things that I think is really important in the AI world is that we approach everything from a human perspective because it's a big disruptive shift, and there's a lot of concern over it. So, if you get people to connect to it in a humanistic way, like coworker Viv works along with coworker Sophia, and Viv has this expertise, Sophia has this expertise, and has better improving ways to interface with people who have names that aren't a lot different from them and have skillsets that aren't a lot different. When you look at the AIS, they don't mind working longer hours. Let them work the weekends so I can spend hours with my family. Let them work the crazy shifts. So, things are different in that regard. But the relationship aspect of how the workplace works, try not to disrupt that too much. >> So, then on a consumption side, with the person coworker that's working with the persona, how do they interact with it, how do they get the data out, and I guess even more importantly, maybe, how do they get the new data back in to continue to train the model? >> So, the biggest thing you have to focus on with a human and machine learning interface that doesn't require a program or a data science, is that the language that the AI is taught in is human language, natural human language. So, we developed a lot of natural human language files that are pretty neat because a human coworker in California here could be interfacing in english to their coworker, and at the same time, someone speaking Mandarin in Shanghai could be interfacing with the same coworker speaking mandarin unless you can get multilingual functionality. Right now, to answer your question, people are doing it in a text based scenario. But the future vision, I think when the industry timing is right, is we view that every one of the coworkers we're developing will have a very distinct unique fingerprint of a voice. So, therefor, when you're engaging with your coworker using voice, you'll begin to recognize, oh, that's Dax, or that's Viv, or that's Sophia, based on their voice. So, like many people, this is how we're communicating with voice, and we believe the same thing's going to occur. And a lot of that's in timing. That's the direction where things are headed. >> Interesting. The whole voice aspect is just a whole 'nother interesting thing in terms of what type of voice personality attributes associated with voice. That's probably going to be a huge piece in terms of the adoption, in terms of having a true coworker experience, if you will. >> One of the things we haven't figure out, and these are important questions, and there's so many unknowns, is we feel really confident that the AI persona should have a unique voice because then I know who I'm engaging with, and I can connect by ear without them saying what their name is. But what does an AI persona look like? That's something where actually we don't know that, and we explore different things and, oh, that looks scary, or oh, that doesn't make sense. Should it look like anything? Which has largely been the approach of what does an Alexa or a Siri look like. As you continue to advance those engagements, and particularly when augmented reality comes into play, through augmented reality, if you're able to look and say, "Oh, a coworker's working over there," there's some value in that. But what is it going to look like? That's interesting, and we don't know that. >> Hopefully, better than those things at the San Jose Airport that are running around. >> Yeah, exactly. >> Classic robot. All right, Scott, very interesting story. I look forward to watching you grow and develop over time. >> Awesome, it's good to talk. >> Absolutely, all right, he's Scott Noteboom, he's from Litbit. I'm Jeff Frick, you're watching The Cube. We're at When IoT met AI: The Intelligence of Things, here at San Jose California. We'll be right back after the short break. Thanks for watching. (upbeat music)

>> Narrator: From the Fairmont Hotel in the heart of Silicon Valley it's theCUBE covering when IoT met AI, The Intelligence of Things. Brought to you by Western Digital. >> Hey welcome back here everybody Jeff Frick here with theCUBE. We're in San Jose, California at the Fairmont Hotel, at the when IoT met AI show, it's all about the intelligence of things. A lot of really interesting start ups here, we're still so early days in most of this technology. Facial recognition gets a lot of play, iris recognition, got to get rid of these stupid passwords. We're really excited to have our next guest, he's Modar Alaoui, he's the CEO and founder of Eyeris. And it says here Modar that you guys are into face analytics and emotion recognition. First off welcome. >> Thank you so much for having me. >> So face analytics, I'm a clear customer I love going to clear at the airport, I put my two fingers down, I think they have my iris, they have different things but what's special about the face compared to some of these other biometric options that people have? >> We go beyond just the biometrics, we do pretty much the entire suites of face analytics. Anything from eye openness, face, gender, emotion recognition, head bows, gaze estimation, et cetera et cetera. So it is pretty much anything and everything you can derive from the face including non verbal clues, yawning, head nod, head shake, et cetera. >> That was a huge range of things, so clearly just the face recognition to know that I am me probably relatively straight forward. A couple anchor points, does everything measure up and match the prior? But emotion that's a whole different thing, not only are there lots of different emotions, but the way I express my emotion might be different than the way you express the very same emotion. Right, everybody has a different smile. So how do you start to figure out the algorithms to sort through this? >> Right, so you're right. There are some nuances between cultures, ages, genders, ethnicities and things like that. Generally they've been universalized for the past three and a half decades by the scholars the psychologists et cetera. So what they actually have a consensus on is that there are only seven or six universal emotions plus neutral. >> Six, what are the six? >> Joy, surprise, anger, disgust, fear, sadness, and neutral. >> Okay and everything is some derivation of that, you can kind of put everything into little buckets. >> That is correct so think of them as seven universal colors or seven primary colors and then everything else is a derivative of that. The other thing is that emotions are hard wired into our brain they happen in a 1/15th or a 1/25th of a second, particularly micro expressions. And they can generally give up a lot of information as to whether a person has suppressed the certain emotion or not or whether they are thinking about something negatively before they could respond positively, et cetera. >> Okay so now you've got the data, you know how I'm feeling, what are you doing with it? It must tie back to all types of different applications I would assume. >> That's right there are a number of applications. Initially when we created this, what we call, enabling technology we wanted to focus on two things. One, is what type of application could have the biggest impact but also the quickest adoption in terms of volumes. Today we focus on driver monitoring AI as well as occupants monitoring AI so we focus on Autonomous and semi autonomous vehicles. And a second application is social robotics, but in essence if you think of a car it's also another robot except that social robotics are those potentially AI engines, or even AI engines in form of an actual robot that communicates with humans. Therefore, the word social. >> Right, so I can see a kind of semi autonomous vehicle or even a not autonomous vehicle you want to know if I'm dosing off. And some of those things have been around in a basic form for a little while. But what about in an autonomous vehicle is impacted by my emotion as a passenger, not necessarily a driver if it's a level five? >> That's right, so when we talk about an autonomous vehicle I think what you're referring to is level five autonomy where a vehicle does not actually have a steering wheel or gas pedal or anything like that. And we don't foresee that those will be on a road for at least another 10 years or more. The focus today is on level two, three, and four, and that's semi autonomy. Even for autonomous, fully autonomous vehicles, you would see them come out with vision sensors or vision AI inside the vehicle. So that these sensors could, together with the software that could analyze everything that's happening inside, cater to the services towards what is going to be the ridership economy. Once the car drives itself autonomously, the focus shifts from the driver to the occupants. As a matter of a fact it's the occupants that would be riding in these vehicles or buying them or sharing them, not the driver. And therefore all these services will revolve around who is inside the vehicle like age, gender emotion, activity, et cetera. >> Interesting, so all these things the age, gender emotion, activity, what is the most important do you think in terms of your business and kind of where as you say you can have a big impact. >> We can group them into two categories, the first one is safety obviously, eye openness, head bows, blinking, yawning, and all these things are utmost importance especially focused on the driver at this point. But then there is a number of applications that relates to comfort and personalization. And so those could potentially take advantage of the emotions and the rest of the analytics. >> Okay, so then where are you guys, Eyeris as a company? Where do have some installations I assume out there? Are you still early days kind of? Where are you in terms of the development of the company? >> We have quite a mature product, what I can disclose is we have plans to go into mass production starting 2018. Some plans for Q4 2017 have been pushed out. So we'll probably start seeing some of those in Q1, Q2 2018. >> Okay. >> We made some announcements earlier this year at CS with Toyota and Honda. But then we'll be seeing some mass volume starting 2019 and beyond. >> Okay, and I assume you're a cloud based solution. >> We do have that as well, but we are particularly a local processing solution. >> Jeff: Oh you are? >> Yes so think of it as an edge computing type of solution. >> Okay and then you work with other peoples sensors and existing systems or are you more of a software component that plugs in? Or you provide the whole system in terms of the, I assume, cameras to watch the people? >> So we're a software company only, we however, are hardware processor camera diagnostic. And of course for everything to succeed there will have to be some components of sensor fusion. And therefore we can work and do work with other sensor companies in order to provide higher confidence level of all the analytics that we provide. >> Pretty exciting, so is it commercially available you're GA now or not quite yet? >> We'll be commercially available, you'll start seeing it on the roads or in the market sometime early next year. >> Sometime early next year? Alright well we will look forward to it. >> Thank you so much. >> Very exciting times, alright, he's Modar Alaoui. And he's going to be paying attention to you to make sure you're paying attention to the roads. So you don't fall asleep, or doze off and go to sleep. So I'm Jeff Frick, you're watching theCUBE at IoT met AI, The Intelligence of Things. San Jose, California, we'll be right back after this short break, thanks for watching. (bright techno music)

(upbeat music) >> Announcer: From the Fairmont Hotel, in the heart of Silicon Valley, it's theCUBE. Covering, When IoT met AI: The Intelligence of Things. Brought to you by Western Digital. >> Welcome back everybody. Jeff Frick here with theCUBE. We're at Downtown San Jose at the Fairmont Hotel at a small little conference, very intimate affair, talking about IoT and AI, The Intelligence of Things. When IoT met AI. Now, they've got a cool little start up, kind of expo hall. We're excited to have our next guest here from that. It's Mike Wilson, he's the CEO of BriteThings. Mike, welcome. >> Good to be here, Jeff, how you doin'? >> Absolutely. So, BriteThings. What are BriteThings? >> BriteThings are intelligent plugs, power strips, wall sockets, anything that fits into the plug load space. It learns users behavior and then provides them an intelligent on-off schedule. The goal here is to turn stuff off when it's on and not being needed. >> Right. >> So wasted energy. Nights and weekends in the workspace, for example. >> It sounds like such a simple thing. >> Totally. >> But we were talking before we turned the cameras on, this actually has giant economic impact >> It does. >> in building maintenance, which is a huge category >> Yup. >> as you said, I'll let you kind of break down the numbers as to where >> Sure. >> that energy's being spent and the impact that you guys are having. >> Well our customers are building owners and operators, and they pay an electrical bill to run that building. It's a cost of running the building. About 27% of it goes to lighting, about 38% goes to heating and cooling, and all the rest goes to plug loads. And where we come to the market it, of course there's huge lighting companies, famous names, same with HVAC, but no one's doing anything about plug loads, and the reason is is because plug loads are distributed, they're hard to control. And so what we bring to the market is a product that is small, inexpensive, and can suddenly give owners and operators all the control that they enjoy with lighting and HVAC over their plug loads. >> So it's kind of like Dest, in that it takes a relatively simple function, now because of the cloud, because of the internet, you can add a lot more intelligence into a relatively, I don't want to say dumb device, but the device itself doesn't have to have that much power 'cause you can put the application somewhere else. >> Exactly, so if you just imagine, you're sitting here with me right now. Probably at your workplace and at home there's a bunch of stuff turned on, you're not using it, >> Right >> but you're spending money to keep it powered up, and that's causing CO2 to be generated at power plant down the road. So that's bad for your pocket, it's bad for the environment. So if we can automatically turn that stuff off, then people don't have to worry about it. We can measure it, so here's where the money is. >> Right. >> Not only energy savings, but data. So I can tell you when you turned your stuff on and off, so that means human presence. When you're at work, there's a value to that. If you're going to put a floor of an office building out there and heat it or light it, we can tell you if people are there or not. So you can look at that and make, and save even more money. >> Jeff: Right. >> We've got one customer that uses our product for inventory management. If it plugs in, you can see it on our screen, and you can see if it's on or off, if it's connected and how it's running. So that kind of data ends up being valuable, not only for energy savings, because we turn stuff on and off, but human presence, inventory control, the list goes on and on. Our customers actually every year are coming up with new ways to use our device. >> Right. And just for the baseline savings, you just basically plug it in and turn it on, and you're reporting some huge savings just by just the basic operation of your strips versus a regular strip. >> Exactly. So just imagine, this device is learning your behavior, so that's part of our, you know, that's kind of our core competency here, is these devices measure the amount of energy you're using. When you're not using something, it goes into standby mode, or sleep mode. Then we turn that off to save you the money. But the way we're able to do that is using artificial intelligence to learn patterns, and take those patterns and you can basically guess the best optimized schedule for your devices to be turned and off. >> Right. >> On and off. So if you imagine you've got 100,000 employees, 100,000 different schedules, this thing has to be smart and it can't affect worker productivity. >> Right. >> So we have to be smart enough to know when to turn it on before you come into work, when to turn it off to save you the max amount of money, and be able to measure all of that so you can roll that up and see how much money you're saving. How much CO2 are you reducing? >> Right. >> You know, so sustainability officers love our product too. >> So do you integrate with other types of intelligent systems in that space? The lightings, and the HVAC? >> Yeah. Exactly. So one of the most important things is, I've got a portfolio, my office building is a portfolio of devices and systems, so just one of them is our plug load management, right? So I want to be able to see my plug load in my current control panel. So we've got APIs where our cloud technology is able to take that reporting and stick it into, for example, a Lucid control panel. We're working with Trane right now to integrate their BACnet solution for their building control management. >> Right, right. >> So that their customers are able to see lighting, HVAC, and plug load, >> Just what I was going to say. >> right off the same old screen and operating tools that they've always used. >> Right, right. What's kind of the typical ROI that you pitch people just for the straight-up money savings that they're going to get? >> We got our foot in the door by saying we can reduce your plug load cost a minimum of 30%, and what we're seeing on average is about 40 to 45%. >> Wow. >> It's a huge huge reduction. >> Now where do you go next? >> Well, conquer the world. (Jeff laughs) You know, so imagine this, anywhere in the commercial office space where there's a plug, so let your mind go, how many power strips are out there? >> Right, right. >> How many of those-- >> We're using about 20 of them right here. >> Yeah, so, just, you know, every person at every desk is a potential customer. Every time there's a coffeemaker or a break room, a fax machine, you know, any piece of equipment that's plugged in, we can save you money. Vending machines. We have a customer with these, you know, raise and lower desks. Crazy, they want to just see, they don't want to save energy, they want to know who's using that and how often. >> Jeff: Right, right. >> Our device can do that, too. >> Right. >> And that's that data I was telling you about. Once you start collecting data of how people use plugged-in devices, I'm collecting information about you, how you use your laptop, how you use your charger, how often. >> Because the signature on the draw is different depending on the activity of the device. >> You got it. Exactly. >> I love this. You know, it's so funny because the second-order impact of all these types of things is so much more significant than people give it credit, I think. >> It's about the data. >> Jeff: Yeah. >> And our customer's just love that, because the data gives them control, and when you have control, cost savings. >> And is it just commercial, or you sell them for regular retail customers as well? Or do you-- >> I imagine some day in the future that's a potential, but you know, our focus right now, 'cause the big problem out there is that buildings use 40% of all the energy generated in the United States, and commercial space is the big opportunity, because nights and weekends. >> Right. >> Stuff should be turned off, and we can do that right now. >> Right, right. >> We're the market doing it. >> Buildings with big, big POs. >> Yup. (Jeff laughs) >> Alright, Michael, sounds like exciting stuff, can't wait til I can get one at Best Buy or Office Depot, or something. >> Coming to a store near you, or www.britethings.com. >> Alright, thanks a lot, he's Mike Wilson. Save some energy, get one of these things when they're available, or at least tell the boss to get one at the office. (Michael laughs) >> Definitely. >> Alright, I'm Jeff Frick, you're watching theCUBE. When IoT meets AI in San Jose, California. Thanks for watching. (upbeat music)

>> Announcer: Live from the Computer History Museum, in the heart of Silicon Valley, it's theCUBE, covering Food IT: Fork to Farm. Brought to you by Western Digital. >> Hi, welcome back, I'm Lisa Martin with theCUBE, we are at the Farm IT event. This is an incredible opportunity to talk with folks that are experts in agriculture, food and agriculture, academia, farmers, producers, those all across the food chain. The theme of this event is Fork to Farm, and I'm excited to be joined by my next two guests, we have Dan Sonke, the Director of Sustainable Agriculture from Campbell's Soup, welcome. >> Thank you. >> And you can't say this, but Dan has Campbell Soup tennis shoes on and they're awesome. And David Sypnieski, the Founder and CEO of Athena Intelligence, welcome gentlemen. >> Thank you. >> Thank you, good to be here. >> So this has been, before we went on we were kind of talking about kind of my thoughts on Ag-Tech, and this is a really interesting and unique opportunity for theCUBE, to really look at the influences of Big Data and Analytics, Cloud Computing, Open-source Software, Blockchain, and how this all can be very influential across the food chain and you know, from the event's theme perspective, it's really been a lot this morning, talking about the tech-enabled food consumer really driving a lot of this change, expectation-wise. But Dan, first question to you, knowing, growing up on Campbell's Soup as a kid, founded in 1869, how is Campbell's Soup taking action to implement not only support-sustainable agriculture, but also, what were the drivers? >> Well, we definitely see consumers driving interest in where the food comes from, where ingredients that go into Campbell's Soup come from. We, a few years ago, decided that we wanted to be a company that makes real food that matters for life's moments, so that's our mission, that's our purpose, and so we want to connect to consumers with the information that supports that claim, that the food is trustworthy, that it's authentic, and that it resonates with the emotional side of how it's consumed in families, and the moments that matter. >> And also probably from a branch perspective, this is a historic brand in the United States, and that's probably quite important to meet those needs. >> Absolutely, we want to we the most transparent food company, we want to be open and honest with our consumers, and satisfy their desire for real food. >> So talk to us about kind of the genesis of the sustainability in agriculture at Campbell, when did that start? And really, besides the consumers, maybe some on the customer side, who was really driving this initiative? >> Well, we drive it internally, so six years ago, we decided to venture into sustainable agriculture in a formal way. We did a stakeholder assessment, so we talked to customers, we talked to investors, we talked to farmers, suppliers, folks inside the company, outside the company, North America, Europe, Australia, and asked them a series of questions, and said where should we focus, what are the crops, what are the subject areas we should focus on in agriculture sustainability? And we came up with a focus on tomatoes and other vegetables that people think of when they think of Campbell's Soup, we're largely a vegetable nutrition, and whole-grain nutrition company, so we wanted to focus there. And we focused on water, fertilizer, greenhouse gases, soil and pesticides, so that was our focus area, and we really took a measure-to-manage approach, so intentionally going to farmers, starting with tomatoes, with a limited set of questions that capture a lot of information and would be information growers would have, so we asked them how much water did you apply to make the crop, how much fertilizer do you use, what was the irrigation system, what are some of the decision tools that you used to make informed decisions? And so we started collecting that data. We also started capturing the geographic locations of the fields, believing that the technology would come to enable us to put that together, and lo and behold, fast-forward five years, now we have five years of data. We've tracked some really great stuff that our farmers have done. For example, last year water use per pound of tomato grown, was down by 20% over our first year of tracking that data. >> Wow. >> Huge gains, and efficiency and, you know, especially since it's a California crop, that was in the period of a five-year drought, so very encouraging to see that growers can do that kind of thing, and very proud of our growers for doing that. >> Absolutely, and on the technology side, so we've got David here. Athena Intelligence, talk to us a little bit about the genesis of Athena Intelligence, and how your working in partnership with Campbell's Soup. >> Sure, so I've got a storied background in agricultural tech work with production, growers, ag-tech companies, processors like Campbell's and others. And several years ago I kind of realized the fact that while all of this technology is from Silicon Valley and around the world, it's starting to, kind of make it's way into agriculture. An assumption that everyone makes is that the data is ready to be used in some sort of technology. >> Right. >> Alright, so kind of the the running joke in the field is that, you know, that a lot of technology has built a lot of solutions that are desperately looking for a problem to solve. And the problem, while it sounds simple, it not so easy to put together. But the problem is that, as Campbell's Soup for example, was collecting all of that data, you know, the entire industry has never really been familiar with the structure of how do you actually use data in any kind of meaningful kind of data science or analytical way and so, just being able to compile it all from various different formats and sources was a burden, so while you had all this data, it actually couldn't be used at all. And so Athena Intelligence was about basically, me coming to the realization, and collaborating with Dan, and Campbell's has been a great partner of saying, you know, we're going to solve that one problem, the unglamourous, the unsexy, problem of building a piece of technology which can efficiently and automatically begin to clean up, and normalize, and standardized data sets from multiple different sources and-- >> And we're talking about like data from weather sources, sensors, satellite imagery-- >> Right, so it's a fusion of public and private data, so the public data, everything from satellite imagery to soil, to weather stations, river flows, 98 different attributes of the weather, and water-related data. And then of course all of the private data, both Campbell's internal processing data, and then all the data that they're collaborating with their suppliers so, it's a pretty broad assortment which comes from, I mean the formats are everything from a hand-written notebook, to a PDF, to Excel to-- >> Wow. >> It's all over the board. >> So this is really Big Data and Analytics, being able to bring and aggregate data from different sources, facilitate data discovery. >> We're making data efficient right now, because the problem is that it's so, it's such a laborious effort. You know, 90% of the time people are putting in, just trying to clean and organize it. >> Right. >> Leaving very little time to be able to analyze it, let alone make any decisions or collaborate on it. So we're addressing that 90% of the time that people spend on trying to put the stuff together in the first place. >> Okay so Dan, walk us through kind of a use-case example of how your implementing, or have implemented, Athena Intelligence software, and what some of the outcomes have been so far. >> Right, so the goal has been to take the quality data that comes in to our systems, and that is one area where we do use data historically quite a bit, we have tons of data on every load of tomatoes that comes into our processing plants. But then we're marrying that data to the publicly available weather, soil, water data, and the data that the growers report on sustainability practices. And the goal is to find the win, win, win, the win for the environment, the win for the farm profitability, and the win for Campbell's Soup quality, and sustainability drivers as well. And the example that we're currently pursuing is tomato solids, so that's an obscure term for most people, but it's a industry measurement of how much sugar is in the tomatoes basically. >> Okay. >> The solids of the tomatoes coming in, affect how they process into our ingredients, the higher solids, the easier they are for us to process, and the less energy it requires for us to do that. So it's a sustainability win as well. We already pay growers for higher solids. We know a few things that can generate higher solids on the farm, but we think there are more pieces of information that have been hiding in that Big Data set. So can we tease out what soils produce higher solids, or what irrigation practices drive higher solids, or whatever it is, so we're in the process right now. We've got a project going between our research innovation fund, Athena, and that's the target that we're going after this summer is to dig into five years of data, and find that win. >> Wow. So it sounds like Athena Intelligence has really enabled Campbell's Soup to become a data-driven company? >> Well, we certainly are a data-driven company, but this is extending the reach of the data outside the four walls of our factory-- >> And also into the farmer, so you're really enabling the farmers to embrace data, evaluate what they have. Have you seen any...? So one of the things we were talking about earlier today, or was being talked about was the labor shortages, as well as attrition. So you mentioned you know, things in ledgers and hard copy. Are you also seeing an influence maybe, that Campbell's having to your farmers, becoming much more, less paper-driven, and maybe more modern in terms of the way that they're collecting and storing data? >> Well, I can't say that we can take credit for that, but we certainly want to be one of the many voices at events such as this one, to be a beacon, calling the industry to solve this problem. David really mentioned it. The challenge is, growers don't have the resources to capture data easily. If they were you know, if that was their mindset, they'd probably be accountants and not farmers right? Farm they have, you know, they're in farming for all the attributes of a farm lifestyle, not a data-capture lifestyle. >> Right. >> So capturing that farm data, and making it easy for them to get the data into systems that they can then use, is one of my passions right? A lot of companies are out there saying, "Oh, we can create a platform that will help Campbell's "get information out of the farms." And I keep telling them, "No, if you create the system "that makes it easier for farmers to use their own data, "to get more efficient and more profitable, "they'll put the data in." >> Okay. >> That's not-- >> So you think that's really where the sweet spot is, and the next step is really-- >> And that's how we drive sustainability. >> Because if they, if the tools can help them with the data to make more informed decisions that's, that's what we want to get out of our sustainability programs, it's not just data for reports say, for Campbell's, it's how do we drive progress on the farm, and we do that by creating the systems that everybody can use more easily. >> Well, it's so neat to hear that a company that so many of us know and have grown up with, has evolved so much to be very focused, and have sustainability really, as a core, and it's also great to know that there are technologists out there that have that Ag-Tech experience, that are enabling companies to leverage the power of Big Data, so gentlemen, I want to thank you so much for stopping by theCUBE and sharing your insights with us, we wish you the best of luck, and look forward to seeing what happens in the next few years. >> Thank you. >> Thank you very much. >> My pleasure. And we want to thank you for watching theCUBE again, I'm Lisa Martin, and we are at the Farm IT event From Fork to Farm, or Food IT event. We will be back with some more great guests, so stick around. (techno music)

(upbeat techno music) >> Collibra is a company that was founded in 2008 right before the so-called modern big data era kicked into high gear. The company was one of the first to focus its business on data governance. Now, historically, data governance and data quality initiatives, they were back office functions, and they were largely confined to regulated industries that had to comply with public policy mandates. But as the cloud went mainstream the tech giants showed us how valuable data could become, and the value proposition for data quality and trust, it evolved from primarily a compliance driven issue, to becoming a linchpin of competitive advantage. But, data in the decade of the 2010s was largely about getting the technology to work. You had these highly centralized technical teams that were formed and they had hyper-specialized skills, to develop data architectures and processes, to serve the myriad data needs of organizations. And it resulted in a lot of frustration, with data initiatives for most organizations, that didn't have the resources of the cloud guys and the social media giants, to really attack their data problems and turn data into gold. This is why today, for example, there's quite a bit of momentum to re-thinking monolithic data architectures. You see, you hear about initiatives like Data Mesh and the idea of data as a product. They're gaining traction as a way to better serve the the data needs of decentralized business users. You hear a lot about data democratization. So these decentralization efforts around data, they're great, but they create a new set of problems. Specifically, how do you deliver, like a self-service infrastructure to business users and domain experts? Now the cloud is definitely helping with that but also, how do you automate governance? This becomes especially tricky as protecting data privacy has become more and more important. In other words, while it's enticing to experiment, and run fast and loose with data initiatives, kind of like the Wild West, to find new veins of gold, it has to be done responsibly. As such, the idea of data governance has had to evolve to become more automated and intelligent. Governance and data lineage is still fundamental to ensuring trust as data. It moves like water through an organization. No one is going to use data that is entrusted. Metadata has become increasingly important for data discovery and data classification. As data flows through an organization, the continuously ability to check for data flaws and automating that data quality, they become a functional requirement of any modern data management platform. And finally, data privacy has become a critical adjacency to cyber security. So you can see how data governance has evolved into a much richer set of capabilities than it was 10 or 15 years ago. Hello and welcome to theCUBE's coverage of Data Citizens made possible by Collibra, a leader in so-called Data intelligence and the host of Data Citizens 2022, which is taking place in San Diego. My name is Dave Vellante and I'm one of the hosts of our program which is running in parallel to Data Citizens. Now at theCUBE we like to say we extract the signal from the noise, and over the next couple of days we're going to feature some of the themes from the keynote speakers at Data Citizens, and we'll hear from several of the executives. Felix Van de Maele, who is the co-founder and CEO of Collibra, will join us. Along with one of the other founders of Collibra, Stan Christiaens, who's going to join my colleague Lisa Martin. I'm going to also sit down with Laura Sellers, she's the Chief Product Officer at Collibra. We'll talk about some of the the announcements and innovations they're making at the event, and then we'll dig in further to data quality with Kirk Haslbeck. He's the Vice President of Data Quality at Collibra. He's an amazingly smart dude who founded Owl DQ, a company that he sold to Collibra last year. Now, many companies they didn't make it through the Hadoop era, you know they missed the industry waves and they became driftwood. Collibra, on the other hand, has evolved its business, they've leveraged the cloud, expanded its product portfolio and leaned in heavily to some major partnerships with cloud providers as well as receiving a strategic investment from Snowflake, earlier this year. So, it's a really interesting story that we're thrilled to be sharing with you. Thanks for watching and I hope you enjoy the program. (upbeat rock music) Last year theCUBE covered Data Citizens, Collibra's customer event, and the premise that we put forth prior to that event was that despite all the innovation that's gone on over the last decade or more with data, you know starting with the Hadoop movement, we had Data lakes, we had Spark, the ascendancy of programming languages like Python, the introduction of frameworks like Tensorflow, the rise of AI, Low Code, No Code, et cetera. Businesses still find it's too difficult to get more value from their data initiatives, and we said at the time, you know maybe it's time to rethink data innovation. While a lot of the effort has been focused on, you more efficiently storing and processing data, perhaps more energy needs to go into thinking about the people and the process side of the equation. Meaning, making it easier for domain experts to both gain insights from data, trust the data, and begin to use that data in new ways, fueling data products, monetization, and insights. Data Citizens 2022 is back and we're pleased to have Felix Van de Maele who is the founder and CEO of Collibra. He's on theCUBE. We're excited to have you Felix. Good to see you again. >> Likewise Dave. Thanks for having me again. >> You bet. All right, we're going to get the update from Felix on the current data landscape, how he sees it why data intelligence is more important now than ever, and get current on what Collibra has been up to over the past year, and what's changed since Data citizens 2021, and we may even touch on some of the product news. So Felix, we're living in a very different world today with businesses and consumers. They're struggling with things like supply chains, uncertain economic trends and we're not just snapping back to the 2010s, that's clear, and that's really true as well in the world of data. So what's different in your mind, in the data landscape of the 2020s, from the previous decade, and what challenges does that bring for your customers? >> Yeah, absolutely, and and I think you said it well, Dave and the intro that, that rising complexity and fragmentation, in the broader data landscape, that hasn't gotten any better over the last couple of years. When when we talk to our customers, that level of fragmentation, the complexity, how do we find data that we can trust, that we know we can use, has only gotten more more difficult. So that trend that's continuing, I think what is changing is that trend has become much more acute. Well, the other thing we've seen over the last couple of years is that the level of scrutiny that organizations are under, respect to data, as data becomes more mission critical, as data becomes more impactful than important, the level of scrutiny with respect to privacy, security, regulatory compliance, as only increasing as well. Which again, is really difficult in this environment of continuous innovation, continuous change, continuous growing complexity, and fragmentation. So, it's become much more acute. And to your earlier point, we do live in a different world and and the past couple of years we could probably just kind of brute force it, right? We could focus on, on the top line, there was enough kind of investments to be, to be had. I think nowadays organizations are focused or are, are, are are, are, are in a very different environment where there's much more focus on cost control, productivity, efficiency, how do we truly get the value from that data? So again, I think it just another incentive for organization to now truly look at data and to scale with data, not just from a a technology and infrastructure perspective, but how do we actually scale data from an organizational perspective, right? You said at the, the people and process, how do we do that at scale? And that's only, only, only becoming much more important, and we do believe that the, the economic environment that we find ourselves in today is going to be catalyst for organizations to really take that more seriously if, if, if you will, than they maybe have in the have in the past. >> You know, I don't know when you guys founded Collibra, if you had a sense as to how complicated it was going to get, but you've been on a mission to really address these problems from the beginning. How would you describe your, your, your mission and what are you doing to address these challenges? >> Yeah, absolutely. We, we started Collibra in 2008. So, in some sense and the, the last kind of financial crisis and that was really the, the start of Collibra, where we found product market fit, working with large financial institutions to help them cope with the increasing compliance requirements that they were faced with because of the, of the financial crisis. And kind of here we are again, in a very different environment of course 15 years, almost 15 years later, but data only becoming more important. But our mission to deliver trusted data for every user, every use case and across every source, frankly, has only become more important. So, what has been an incredible journey over the last 14, 15 years, I think we're still relatively early in our mission to again, be able to provide everyone, and that's why we call it Data Citizens, we truly believe that everyone in the organization should be able to use trusted data in an easy, easy matter. That mission is is only becoming more important, more relevant. We definitely have a lot more work ahead of us because we still relatively early in that, in that journey. >> Well that's interesting, because you know, in my observation it takes 7 to 10 years to actually build a company, and then the fact that you're still in the early days is kind of interesting. I mean, you, Collibra's had a good 12 months or so since we last spoke at Data Citizens. Give us the latest update on your business. What do people need to know about your current momentum? >> Yeah, absolutely. Again, there's a lot of tailwind organizations that are only maturing their data practices and we've seen that kind of transform or influence a lot of our business growth that we've seen, broader adoption of the platform. We work at some of the largest organizations in the world with its Adobe, Heineken, Bank of America and many more. We have now over 600 enterprise customers, all industry leaders and every single vertical. So it's, it's really exciting to see that and continue to partner with those organizations. On the partnership side, again, a lot of momentum in the org in the, in the market with some of the cloud partners like Google, Amazon, Snowflake, Data Breaks, and and others, right? As those kind of new modern data infrastructures, modern data architectures, are definitely all moving to the cloud. A great opportunity for us, our partners, and of course our customers, to help them kind of transition to the cloud even faster. And so we see a lot of excitement and momentum there. We did an acquisition about 18 months ago around data quality, data observability, which we believe is an enormous opportunity. Of course data quality isn't new but I think there's a lot of reasons why we're so excited about quality and observability now. One, is around leveraging AI machine learning again to drive more automation. And a second is that those data pipelines, that are now being created in the cloud, in these modern data architecture, architectures, they've become mission critical. They've become real time. And so monitoring, observing those data pipelines continuously, has become absolutely critical so that they're really excited about, about that as well. And on the organizational side, I'm sure you've heard the term around kind of data mesh, something that's gaining a lot of momentum, rightfully so. It's really the type of governance that we always believed in. Federated, focused on domains, giving a lot of ownership to different teams. I think that's the way to scale data organizations, and so that aligns really well with our vision and from a product perspective, we've seen a lot of momentum with our customers there as well. >> Yeah, you know, a couple things there. I mean, the acquisition of OwlDQ, you know Kirk Haslbeck and, and their team. It's interesting, you know the whole data quality used to be this back office function and and really confined to highly regulated industries. It's come to the front office, it's top of mind for Chief Data Officers. Data mesh, you mentioned you guys are a connective tissue for all these different nodes on the data mesh. That's key. And of course we see you at all the shows. You're, you're a critical part of many ecosystems and you're developing your own ecosystem. So, let's chat a little bit about the, the products. We're going to go deeper into products later on, at Data Citizens 22, but we know you're debuting some, some new innovations, you know, whether it's, you know, the the under the covers in security, sort of making data more accessible for people, just dealing with workflows and processes, as you talked about earlier. Tell us a little bit about what you're introducing. >> Yeah, absolutely. We we're super excited, a ton of innovation. And if we think about the big theme and like, like I said, we're still relatively early in this, in this journey towards kind of that mission of data intelligence that really bolts and compelling mission. Either customers are still start, are just starting on that, on that journey. We want to make it as easy as possible for the, for organization to actually get started, because we know that's important that they do. And for our organization and customers, that have been with us for some time, there's still a tremendous amount of opportunity to kind of expand the platform further. And again to make it easier for, really to, to accomplish that mission and vision around that Data Citizen, that everyone has access to trustworthy data in a very easy, easy way. So that's really the theme of a lot of the innovation that we're driving, a lot of kind of ease of adoption, ease of use, but also then, how do we make sure that, as clear becomes this kind of mission critical enterprise platform, from a security performance, architecture scale supportability, that we're truly able to deliver that kind of an enterprise mission critical platform. And so that's the big theme. From an innovation perspective, from a product perspective, a lot of new innovation that we're really excited about. A couple of highlights. One, is around data marketplace. Again, a lot of our customers have plans in that direction, How to make it easy? How do we make How do we make available to true kind of shopping experience? So that anybody in the organization can, in a very easy search first way, find the right data product, find the right dataset, that they can then consume. Usage analytics, how do you, how do we help organizations drive adoption? Tell them where they're working really well and where they have opportunities. Homepages again to, to make things easy for, for people, for anyone in your organization, to kind of get started with Collibra. You mentioned Workflow Designer, again, we have a very powerful enterprise platform, one of our key differentiators is the ability to really drive a lot of automation through workflows. And now we provided a, a new Low-Code, No-Code kind of workflow designer experience. So, so really customers can take it to the next level. There's a lot more new product around Collibra protect, which in partnership with Snowflake, which has been a strategic investor in Collibra, focused on how do we make access governance easier? How do we, how do we, how are we able to make sure that as you move to the cloud, things like access management, masking around sensitive data, PIA data, is managed as a much more effective, effective rate. Really excited about that product. There's more around data quality. Again, how do we, how do we get that deployed as easily, and quickly, and widely as we can? Moving that to the cloud has been a big part of our strategy. So, we launch our data quality cloud product, as well as making use of those, those native compute capabilities and platforms, like Snowflake, Databricks, Google, Amazon, and others. And so we are bettering a capability, a capability that we call push down, so we're actually pushing down the computer and data quality, to monitoring into the underlying platform, which again from a scale performance and ease of use perspective, is going to make a massive difference. And then more broadly, we talked a little bit about the ecosystem. Again, integrations, we talk about being able to connect to every source. Integrations are absolutely critical, and we're really excited to deliver new integrations with Snowflake, Azure and Google Cloud storage as well. So that's a lot coming out, the team has been work, at work really hard, and we are really really excited about what we are coming, what we're bringing to market. >> Yeah, a lot going on there. I wonder if you could give us your, your closing thoughts. I mean, you you talked about, you know, the marketplace, you know you think about Data Mesh, you think of data as product, one of the key principles, you think about monetization. This is really different than what we've been used to in data, which is just getting the technology to work has been, been so hard. So, how do you see sort of the future and, you know give us the, your closing thoughts please? >> Yeah, absolutely. And, and I think we we're really at a pivotal moment and I think you said it well. We, we all know the constraint and the challenges with data, how to actually do data at scale. And while we've seen a ton of innovation on the infrastructure side, we fundamentally believe that just getting a faster database is important, but it's not going to fully solve the challenges and truly kind of deliver on the opportunity. And that's why now is really the time to, deliver this data intelligence vision, this data intelligence platform. We are still early, making it as easy as we can, as kind of our, as our mission. And so I'm really, really excited to see what we, what we are going to, how the marks are going to evolve over the next, next few quarters and years. I think the trend is clearly there. We talked about Data Mesh, this kind of federated approach focus on data products, is just another signal that we believe, that a lot of our organization are now at the time, they're understanding need to go beyond just the technology. I really, really think about how to actually scale data as a business function, just like we've done with IT, with HR, with sales and marketing, with finance. That's how we need to think about data. I think now is the time, given the economic environment that we are in, much more focus on control, much more focus on productivity, efficiency, and now is the time we need to look beyond just the technology and infrastructure to think of how to scale data, how to manage data at scale. >> Yeah, it's a new era. The next 10 years of data won't be like the last, as I always say. Felix, thanks so much. Good luck in, in San Diego. I know you're going to crush it out there. >> Thank you Dave. >> Yeah, it's a great spot for an in-person event and and of course the content post-event is going to be available at collibra.com and you can of course catch theCUBE coverage at theCUBE.net and all the news at siliconangle.com. This is Dave Vellante for theCUBE, your leader in enterprise and emerging tech coverage. (upbeat techno music)

>> We're back at the ARIA Las Vegas. We're covering CrowdStrike's Fal.Con 22. First one since 2019. Dave Vellante and Dave Nicholson on theCUBE. Adam Meyers is here, he is the Senior Vice President of Intelligence at CrowdStrike. Adam, thanks for coming to theCUBE. >> Thanks for having me. >> Interesting times, isn't it? You're very welcome. Senior Vice President of Intelligence, tell us what your role is. >> So I run all of our intelligence offerings. All of our analysts, we have a couple hundred analysts that work at CrowdStrike tracking threat actors. There's 185 threat actors that we track today. We're constantly adding more of them and it requires us to really have that visibility and understand how they operate so that we can inform our other products: our XDR, our Cloud Workload Protections and really integrate all of this around the threat actor. >> So it's that threat hunting capability that CrowdStrike has. That's what you're sort of... >> Well, so think of it this way. When we launched the company 11 years ago yesterday, what we wanted to do was to tell customers, to tell people that, well, you don't have a malware problem, you have an adversary problem. There are humans that are out there conducting these attacks, and if you know who they are what they're up to, how they operate then you're better positioned to defend against them. And so that's really at the core, what CrowdStrike started with and all of our products are powered by intelligence. All of our services are our OverWatch and our Falcon complete, all powered by intelligence because we want to know who the threat actors are and what they're doing so we can stop them. >> So for instance like you can stop known malware. A lot of companies can stop known malware, but you also can stop unknown malware. And I infer that the intelligence is part of that equation, is that right? >> Absolutely. That that's the outcome. That's the output of the intelligence but I could also tell you who these threat actors are, where they're operating out of, show you pictures of some of them, that's the threat intel. We are tracking down to the individual persona in many cases, these various threats whether they be Chinese nation state, Russian threat actors, Iran, North Korea, we track as I said, quite a few of these threats. And over time, we develop a really robust deep knowledge about who they are and how they operate. >> Okay. And we're going to get into some of that, the big four and cyber. But before we do, I want to ask you about the eCrime index stats, the ECX you guys call it a little side joke for all your nerds out there. Maybe you could explain that Adam >> Assembly humor. >> Yeah right, right. So, but, what is that index? You guys, how often do you publish it? What are you learning from that? >> Yeah, so it was modeled off of the Dow Jones industrial average. So if you look at the Dow Jones it's a composite index that was started in the late 1800s. And they took a couple of different companies that were the industrial component of the economy back then, right. Textiles and railroads and coal and steel and things like that. And they use that to approximate the overall health of the economy. So if you take these different stocks together, swizzle 'em together, and figure out some sort of number you could say, look, it's up. The economy's doing good. It's down, not doing so good. So after World War II, everybody was exuberant and positive about the end of the war. The DGI goes up, the oil crisis in the seventies goes down, COVID hits goes up, sorry, goes down. And then everybody realizes that they can use Amazon still and they can still get the things they need goes back up with the eCrime index. We took that approach to say what is the health of the underground economy? When you read about any of these ransomware attacks or data extortion attacks there are criminal groups that are working together in order to get things spammed out or to buy credentials and things like that. And so what the eCrime index does is it takes 24 different observables, right? The price of a ransom, the number of ransom attacks, the fluctuation in cryptocurrency, how much stolen material is being sold for on the underground. And we're constantly computing this number to understand is the eCrime ecosystem healthy? Is it thriving or is it under pressure? And that lets us understand what's going on in the world and kind of contextualize it. Give an example, Microsoft on patch Tuesday releases 56 vulnerabilities. 11 of them are critical. Well guess what? After hack Tuesday. So after patch Tuesday is hack Wednesday. And so all of those 11 vulnerabilities are exploitable. And now you have threat actors that have a whole new array of weapons that they can deploy and bring to bear against their victims after that patch Tuesday. So that's hack Wednesday. Conversely we'll get something like the colonial pipeline. Colonial pipeline attack May of 21, I think it was, comes out and all of the various underground forums where these ransomware operators are doing their business. They freak out because they don't want law enforcement. President Biden is talking about them and he's putting pressure on them. They don't want this ransomware component of what they're doing to bring law enforcement, bring heat on them. So they deplatform them. They kick 'em off. And when they do that, the ransomware stops being as much of a factor at that point in time. And the eCrime index goes down. So we can look at holidays, and right around Thanksgiving, which is coming up pretty soon, it's going to go up because there's so much online commerce with cyber Monday and such, right? You're going to see this increase in online activity; eCrime actors want to take advantage of that. When Christmas comes, they take vacation too; they're going to spend time with their families, so it goes back down and it stays down till around the end of the Russian Orthodox Christmas, which you can probably extrapolate why that is. And then it goes back up. So as it's fluctuating, it gives us the ability to really just start tracking what that economy looks like. >> Realtime indicator of that crypto. >> I mean, you talked about, talked about hack Wednesday, and before that you mentioned, you know, the big four, and I think you said 185 threat actors that you're tracking, is 180, is number 185 on that list? Somebody living in their basement in their mom's basement or are the resources necessary to get on that list? Such that it's like, no, no, no, no. this is very, very organized, large groups of people. Hollywood would have you believe that it's guy with a laptop, hack Wednesday, (Dave Nicholson mimics keyboard clacking noises) and everything done. >> Right. >> Are there individuals who are doing things like that or are these typically very well organized? >> That's a great question. And I think it's an important one to ask and it's both it tends to be more, the bigger groups. There are some one-off ones where it's one or two people. Sometimes they get big. Sometimes they get small. One of the big challenges. Have you heard of ransomware as a service? >> Of course. Oh my God. Any knucklehead can be a ransomwarist. >> Exactly. So we don't track those knuckleheads as much unless they get onto our radar somehow, they're conducting a lot of operations against our customers or something like that. But what we do track is that ransomware as a service platform because the affiliates, the people that are using it they come, they go and, you know, it could be they're only there for a period of time. Sometimes they move between different ransomware services, right? They'll use the one that's most useful for them that that week or that month, they're getting the best rate because it's rev sharing. They get a percentage that platform gets percentage of the ransom. So, you know, they negotiate a better deal. They might move to a different ransomware platform. So that's really hard to track. And it's also, you know, I think more important for us to understand the platform and the technology that is being used than the individual that's doing it. >> Yeah. Makes sense. Alright, let's talk about the big four. China, Iran, North Korea, and Russia. Tell us about, you know, how you monitor these folks. Are there different signatures for each? Can you actually tell, you know based on the hack who's behind it? >> So yeah, it starts off, you know motivation is a huge factor. China conducts espionage, they do it for diplomatic purposes. They do it for military and political purposes. And they do it for economic espionage. All of these things map to known policies that they put out, the Five Year Plan, the Made in China 2025, the Belt and Road Initiative, it's all part of their efforts to become a regional and ultimately a global hegemon. >> They're not stealing nickels and dimes. >> No they're stealing intellectual property. They're stealing trade secrets. They're stealing negotiation points. When there's, you know a high speed rail or something like that. And they use a set of tools and they have a set of behaviors and they have a set of infrastructure and a set of targets that as we look at all of these things together we can derive who they are by motivation and the longer we observe them, the more data we get, the more we can get that attribution. I could tell you that there's X number of Chinese threat groups that we track under Panda, right? And they're associated with the Ministry of State Security. There's a whole other set. That's too associated with the People's Liberation Army Strategic Support Force. So, I mean, these are big operations. They're intelligence agencies that are operating out of China. Iran has a different set of targets. They have a different set of motives. They go after North American and Israeli businesses right now that's kind of their main operation. And they're doing something called hack and lock and leak. With a lock and leak, what they're doing is they're deploying ransomware. They don't care about getting a ransom payment. They're just doing it to disrupt the target. And then they're leaking information that they steal during that operation that brings embarrassment. It brings compliance, regulatory, legal impact for that particular entity. So it's disruptive >> The chaos creators that's.. >> Well, you know I think they're trying to create a they're trying to really impact the legitimacy of some of these targets and the trust that their customers and their partners and people have in them. And that is psychological warfare in a certain way. And it, you know is really part of their broader initiative. Look at some of the other things that they've done they've hacked into like the missile defense system in Israel, and they've turned on the sirens, right? Those are all things that they're doing for a specific purpose, and that's not China, right? Like as you start to look at this stuff, you can start to really understand what they're up to. Russia very much been busy targeting NATO and NATO countries and Ukraine. Obviously the conflict that started in February has been a huge focus for these threat actors. And then as we look at North Korea, totally different. They're doing, there was a major crypto attack today. They're going after these crypto platforms, they're going after DeFi platforms. They're going after all of this stuff that most people don't even understand and they're stealing the crypto currency and they're using it for revenue generation. These nuclear weapons don't pay for themselves, their research and development don't pay for themselves. And so they're using that cyber operation to either steal money or steal intelligence. >> They need the cash. Yeah. >> Yeah. And they also do economic targeting because Kim Jong Un had said back in 2016 that they need to improve the lives of North Koreans. They have this national economic development strategy. And that means that they need, you know, I think only 30% of North Korea has access to reliable power. So having access to clean energy sources and renewable energy sources, that's important to keep the people happy and stop them from rising up against the regime. So that's the type of economic espionage that they're conducting. >> Well, those are the big four. If there were big five or six, I would presume US and some Western European countries would be on there. Do you track, I mean, where United States obviously has you know, people that are capable of this we're out doing our thing, and- >> So I think- >> That defense or offense, where do we sit in this matrix? >> Well, I think the big five would probably include eCrime. We also track India, Pakistan. We track actors out of Columbia, out of Turkey, out of Syria. So there's a whole, you know this problem is getting worse over time. It's proliferating. And I think COVID was also, you know a driver there because so many of these countries couldn't move human assets around because everything was getting locked down. As machine learning and artificial intelligence and all of this makes its way into the cameras at border and transfer points, it's hard to get a human asset through there. And so cyber is a very attractive, cheap and deniable form of espionage and gives them operational capabilities, not, you know and to your question about US and other kind of five I friendly type countries we have not seen them targeting our customers. So we focus on the threats that target our customers. >> Right. >> And so, you know, if we were to find them at a customer environment sure. But you know, when you look at some of the public reporting that's out there, the malware that's associated with them is focused on, you know, real bad people, and it's, it's physically like crypted to their hard drive. So unless you have sensor on, you know, an Iranian or some other laptop that might be target or something like that. >> Well, like Stuxnet did. >> Yeah. >> Right so. >> You won't see it. Right. See, so yeah. >> Well Symantec saw it but way back when right? Back in the day. >> Well, I mean, if you want to go down that route I think it actually came from a company in the region that was doing the IR and they were working with Symantec. >> Oh, okay. So, okay. So it was a local >> Yeah. I think Crisis, I think was the company that first identified it. And then they worked with Symantec. >> It Was, they found it, I guess, a logic controller. I forget what it was. >> It was a long time ago, so I might not have that completely right. >> But it was a seminal moment in the industry. >> Oh. And it was a seminal moment for Iran because you know, that I think caused them to get into cyber operations. Right. When they realized that something like that could happen that bolstered, you know there was a lot of underground hacking forums in Iran. And, you know, after Stuxnet, we started seeing that those hackers were dropping their hacker names and they were starting businesses. They were starting to try to go after government contracts. And they were starting to build training offensive programs, things like that because, you know they realized that this is an opportunity there. >> Yeah. We were talking earlier about this with Shawn and, you know, in the nuclear war, you know the Cold War days, you had the mutually assured destruction. It's not as black and white in the cyber world. Right. Cause as, as Robert Gates told me, you know a few years ago, we have a lot more to lose. So we have to be somewhat, as the United States, careful as to how much of an offensive posture we take. >> Well here's a secret. So I have a background on political science. So mutually assured destruction, I think is a deterrent strategy where you have two kind of two, two entities that like they will destroy each other if they so they're disinclined to go down that route. >> Right. >> With cyber I really don't like that mutually assured destruction >> That doesn't fit right. >> I think it's deterrents by denial. Right? So raising the cost, if they were to conduct a cyber operation, raising that cost that they don't want to do it, they don't want to incur the impact of that. Right. And think about this in terms of a lot of people are asking about would China invade Taiwan. And so as you look at the cost that that would have on the Chinese military, the POA, the POA Navy et cetera, you know, that's that deterrents by denial, trying to, trying to make the costs so high that they don't want to do it. And I think that's a better fit for cyber to try to figure out how can we raise the cost to the adversary if they operate against our customers against our enterprises and that they'll go someplace else and do something else. >> Well, that's a retaliatory strike, isn't it? I mean, is that what you're saying? >> No, definitely not. >> It's more of reducing their return on investment essentially. >> Yeah. >> And incenting them- disincening them to do X and sending them off somewhere else. >> Right. And threat actors, whether they be criminals or nation states, you know, Bruce Lee had this great quote that was "be like water", right? Like take the path of least resistance, like water will. Threat actors do that too. So, I mean, unless you're super high value target that they absolutely have to get into by any means necessary, then if you become too hard of a target, they're going to move on to somebody that's a little easier. >> Makes sense. Awesome. Really appreciate your, I could, we'd love to have you back. >> Anytime. >> Go deeper. Adam Myers. We're here at Fal.Con 22, Dave Vellante, Dave Nicholson. We'll be right back right after this short break. (bouncy music plays)

(upbeat music) >> Hi everyone, and welcome back to managing risk across your extended attack service area with Armis Asset Intelligence Platform. I'm John Furrier, your host. We're here with the CISO Perspective. Alex Schuchman, who is the CISO of Colgate-Palmolive Company. Alex, thanks for coming on. >> Thanks for having me. >> You know, unified visibility across the enterprise service area is about knowing what you got to protect. You can't protect what you can't see. Tell me more about how you guys are able to centralize your view with network assets with Armis. >> Yeah, I think the most important part of any security program is really visibility. And that's one of the building blocks when you're building a security program. You need to understand what's in your environment, what you can control, what is being introduced new into the environment, and that's really what, any solution that gives you full visibility to your infrastructure, to your environment, to all the assets that are there, that's really one of your bread and butter pieces to your security program. >> What's been the impact on your business? >> You know, I think from an IT point of view, running the security program, you know, our key thing is really enabling the business to do their job better. So if we can give them visibility into all the assets that are available in their individual environments, and we're doing that in an automated fashion with no manual collection, you know, that's yet another thing that they don't have to worry about, and then we're delivering. Because really IT is an enabler for the business. And then they can focus really on what their job is, which is to deliver product. >> Yeah, and a lot of changes in their network. You got infrastructure, you got IOT devices, OT devices. So vulnerability management becomes more important. It's been around for a while, but it's not just IT devices anymore. There are gaps in vulnerability across the OT network. What can you tell us about Colgate's use of Armis' vulnerability management? What can you see now? What couldn't you see before? Can you share your thoughts on this? >> Yeah, I think what's really interesting about the kind of manufacturing environments today is, if you look back a number of years, most of the manufacturing equipment was really disconnected from the internet. It was really running in silos. So it was very easy to protect equipment that isn't internet-connected. You could put a firewall, you could segment it off. And it was really on an island on its own. Nowadays, you have a lot of IOT devices. you have a lot of internet-connected devices, sensors providing information to multiple different suppliers or vendor solutions. And you have to really then open up your ecosystem more, which, of course, means you have to change your security posture, and you really have to embrace if there's a vulnerability with one of those suppliers then how do you mitigate the risk associated to that vulnerability? Armis really helps us get a lot of information so that we can then make a decision with our business teams. >> That whole operational aspect of criticality is huge, on the assets knowing what's key. How has that changed the security workload for you guys? >> You know, for us, I mean, it's all about being efficient. If we can have the visibility across our manufacturing environments, then my team can easily consume that information. You know, if we spend a lot of time trying to digest the information, trying to process it, trying to prioritize it, that really hurts our efficiency as a team or as a function. What we really like is being able to use technology to help us do that work. We're not an IT shop. We're a manufacturing shop, but we're a very technical shop so we like to drive everything through automation and not be a bottleneck for any of the actions that take place. >> You know the old expression, is the juice worth the squeeze? It comes up a lot when people are buying tools around vulnerability management, and point for all this stuff. So SaaS solution is key with no agents to deploy. They have that. Talk about how you operationalize Armis in your environment. How quickly did it achieve time to value? Take us through that consumption of the product, and what was the experience like? >> Yeah, I'll definitely say in the security ecosystem, that's one of the biggest promises you hear across the industry. And when we started with Armis, we started with a very small deployment, and we wanted to make sure if it was really worth the lift, to your point. We implemented the first set of plants very quickly, actually even quicker than we had put in our project plan, which is not typical for implementing complex security solutions. And then we were so successful with that, we expanded to cover more of our manufacturing plants, and we were able to get really true visibility across our entire manufacturing organization in the first year, with the ability to also say that we extended that information, that visibility to our manufacturing organization, and they could also consume it just as easily as we could. >> That's awesome. How many assets did you guys discover? Just curious on the numbers? >> Oh, that's the really interesting part. You know, before we started this project we would've had to do a manual audit of our plants, which is typical in our industry. You know, when we started this project and we put in estimates, we really didn't have a great handle on what we were going to find. And what's really nice about the Armis solution is it's truly giving you full visibility. So you're actually seeing, besides the servers, and the PLCs, and all the equipment that you're familiar with, you're also connecting it to your wireless access points. You're connecting it to see any of those IOT devices as well. And then you're really getting full visibility through all the integrations that they offer. You're amazed how many devices you're actually seeing across your entire ecosystem. >> It's like Google maps for your infrastructure. You know, the street view. You want to look at it. You get the, you know, fake tree in there, whatever, but it gives you the picture. That's key. >> Correct. And with a nice visualization and an easy search engine, similar to your Google analogy, you know, everything is really at your fingertips. If you want to find something, you just go to the search bar, click a couple entries and boom, you get your list of the associated devices or the the associated locations devices. >> Well, Alex, I appreciate your time. I know you're super busy at CSIG a lot of your plate. Thanks for coming on sharing. Appreciate it. >> No problem, John. Thanks for having me. >> Okay. In a moment, Bryan Inman, a sales engineer at Armis will be joining me. You're watching theCUBE, the leader in high tech coverage. Thanks for watching. (upbeat music)

(upbeat music) >> Welcome to this Cube Conversation about women in tech and women in cybersecurity, two things I'm very passionate about. Lisa Martin here, with two guests, Debbie Briggs joins us, the Area Vice President, and Chief Security Officer at NETSCOUT, and Tyler Cohen Wood is here as well, the Founder and CEO of MyConnectedHealth. Ladies, it's an honor to have you on the program. I'm excited to talk to you. >> Thank you so much for having us. >> Completely agree. Tyler and I talked a couple of minutes last week and she has a lot to offer to this. >> I know, I was looking at both of your backgrounds. Very impressive. Tyler, starting with you. I see that you are a nationally recognized Cybersecurity Intelligence, National Security Expert, and former Director of Cyber Risk Management for AT&T. And I also saw that you just won a Top 50 Women in Tech Influencers to Follow for 2021 Award. Congratulations, that's amazing. I would love to know way back in the day, how did you even first become interested in tech? >> Well, it was kind of inevitable that I would go into something like tech because as a kid, I was kind of nerdy. I was obsessed with "Star Trek". I would catalog my "Star Trek" tapes by Stardate. I was just really into it. But when I was in college, I mean, it was the late 90's. Cybersecurity just really wasn't a thing. So I went into music and I worked for a radio station. I loved it, but the format of the radio station changed and I wanted to do something different. And I thought, well, computers. I'll move to San Francisco, and I'm sure I can get a job, 'cause they were hiring anyone with a brain, 'cause it was really the dot com boom. And that's really how I got into it. It was just kind of one of those things. (laughs) >> Did you have, was it like network connection, going from music to tech is quite a jump? >> It's a huge jump. It was, but you know, I was young. I was still fresh out of school. I was really interested in learning and I really wanted to get involved in cyber in some capacity, because I became really fascinated with it. So it was just kind of one of those things, that just sort of happened. >> What an interesting talk about a zig-zaggy path. That's a very, very interesting one. And I have to talk about music with you later. That would be interesting. And Debbie, you also have, as Tyler does, 20 years plus experience in cybersecurity. You've been with NETSCOUT since '04. Were you always interested in tech? Did you study engineering or computer science in school, Debbie? >> Yeah, so I think my interest in tech, just like Tyler started at a very young age. I was always interested in how things worked and how people worked. And some day over a drink, I will tell you some funny stories about things I took apart in my parents house, to figure out how it worked. (Lisa and Tyler laughing) They still don't know it. So I guess I- >> I love that. >> I just love that putting it back together, but I took a more traditional route than Tyler did. I do have a degree in Computer Science, went to school a little bit earlier than Tyler. What I would say is, when I was in college, the Computer Science Center was in the basement of the library and we had these really tiny windows and they sort of hit you in the dark. And I think it was my senior year and I went, "I don't want to sit in a room by myself and write code all day and talk to no one." So, you know, I'm a senior and I'm like, "Okay, I got to, this is not, I did not want to write code all day." And so I happened to fall into a great company and moved onto PCs. And from there went to messaging, to networking and into that, I fell into cybersecurity. So I took that more traditional route and I think I've done every job in IT, except for programming, which is what I really got my degree in. >> But you realized early on, you know, "I don't quite think this is for me." And that's an important thing for anybody in any career, to really listen to your gut. It's telling you something. I love how you both got into cybersecurity, which is now, especially in the last 18 months, with what we've seen with the threat landscape, such an incredible opportunity for anyone. But I'd like to know there's not a lot of women in tech, as we know we've been talking about this for a long time now. We've got maybe a quarter of women at the technology roles are filled by women. Tyler, talk to me about some of the challenges that you faced along your journey to get where you are today. >> Well, I mean, you know, like I said, when I started, it was like 1999, 2000. And there were even less women in cybersecurity and in these tech roles than there are now. And you know, it was difficult because, you know, I remember at my first job, I was so interested in learning about Unix and I would learn everything, I read everything about it. And I ended up getting promoted over all of my male colleagues. And you know, it was really awkward because there was the assumption, they would just say things like, "Oh, well you got that because you're a woman." And that was not the case, but it's that type of stereotyping, you know, that we've had to deal with in this industry. Now I do believe that is changing. And I've seen a lot of evidence of that. We're getting there, but we're not there yet. >> And I agree. I agree completely with what Tyler said. You know, when I started, you were the only woman in the room, you got promoted over your male counterparts. You know, I would say even 10 years ago, you know, someone was like, "Well, you could go for any CISCO role and you'd get the job because you're a woman." And I've had to go and say, "No, I might get an interview because I'm a woman, but you don't get the job just because, you know, you check a box." You know, some of that is still out there, but Tyler you're right, things are changing. I think, you know, three things that we all need to focus in on to continue to move us forward and get more women into tech is the first thing is we have to start younger. I think by high school, a lot of girls and young women have been turned off by technology. So maybe, we need to start in the middle school and ensuring that we've got young girls interested. The second thing is, is we have to have mentors. And I always say, if you're in the security industry, you have to turn around and help the next person out. And if that person is a woman, that's great, but we have to mentor others. And it can be young girls, it could be young gentlemen, but we need to mentor that next group up. And you know, if you're in the position to offer internships during the summer, we don't have to stay to the traditional role and go, "Oh, let me hire just intern from the you know IT, they're getting degrees in IT." You can get creative. And my best worker right now was an intern that worked for me, was an intern for me six years ago. And she has a degree in Finance, so nontraditional route into cyber security. And the third thing I think we need to do is, is there things the industry could do to change things and make things, I don't want to say even 'cause they're not uneven, but for example, I forget what survey it was, but if a woman reads a job description and I can do half of it, I'm not going to apply because I don't feel I'll qualify, where men, on the other hand, if they can do three out of ten they'll apply. So do we need to look at the way we write job descriptions, and use different words, you know, rather than must have these skills. You know, sort of leave it a little bit open, like here are the skills we'd like you to have, or have, you know, a handful of the following. So soften some of those job descriptions. And the second thing is once we get women in, we have to be a little bit more, I'll say inclusive. So, if you're a high tech company, look at, you know, your sales organization. When you go to big shows, do you pay more attention to men on the floor than women on the floor? If you have a sales event where you get different customers together, is it a golf outing or is it something that's maybe a little bit more inclusive than just male? So those are the three things I think as an industry we have to focus in on, start younger, get them, you know, work on mentorships specifically in cyber, and the third thing is, look at some of the things that we're doing, as companies both in our HR and sales practices. >> That's a great, that last piece of advice, Debbie is fantastic. That's one that I hadn't thought about, but you're right. If a job description is written, for must have all of these things and a woman that goes, "I only got three out of the ten. I'm not going to even get past, you know, the recruiter here." How can we write things differently? I also loved your idea of bringing in people with diverse backgrounds. I've been in marketing for 16 years and I've met very few people that actually have marketing degrees, a lot of people. So you get that diversity of thought. Tyler, what are some of your thoughts about how we can help expand the role of women in technology? Do you agree with some of the things that Debbie said? >> I love what Debbie said. I agree 100%. And I started laughing because I was thinking about all the golf outings that I've been on and I don't play golf. (all laughing) I think that there is an untapped resource because there's a lot of women who are now interested in changing their careers and that's a big pool of people. And I think that making it more accessible and making it so that people understand what the different cyber security or cyber jobs are, because a lot of people just assume that it's coding, or it's, you know, working on AI, but that's not necessarily true. I mean, there's so many different avenues. There's marketing, there's forensics, there's incident response. I mean, I could go on and on and on. And oftentimes if people don't know that these types of jobs exist, they're not even going to look for them. So making that more well-known, what the different types of opportunities are to people, I think that that would help kind of open more doors. >> And that goes along beautifully with what Debbie was talking about with respect to mentorship. And I would even add sponsorship in there, but becoming a sponsor of a younger female, who's maybe considering tech or is already in tech to help her navigate the career. Look for the other opportunities. Tyler, as you mentioned, there's a lot to cybersecurity, that is beyond coding and AI for example. So maybe getting the awareness out there more. Did either of you have sponsors when you were early in your career? Are you a sponsor now? Debbie, let's start with you. >> So, I'll answer your first question. I guess I was really fortunate that my first job out of college, I had an internship and I happened to have a female boss. And so, although we may not have called it sponsorship or mentor, she taught me and showed me that, you know, women can be leaders. And she always believed in us and always pushed us to do things beyond what we may have thought we were capable of. Throughout the years, someone once told me that we should all have our own personal board of directors. You know, a group of people that when we're making a decision, that may be life-changing or we're unsure, rather than just having one mentor, having a group of people that you, that you know, they don't have to be in cybersecurity. Yeah, I want someone that's on my board of directors that maybe, is a specialist in cybersecurity, but having other executives in other companies, that can also give you that perspective. You know, so I've always had a personal board of directors. I think I've had three or four different mentors. Some of them, I went out and found. Some of them I have joined organizations that have been fortunate enough to become not only a mentor, but a mentee. And I've kept those relationships up over three or four years. And all those people are now on my personal board of directors, that, you know, if I have a life-changing question, I've got a group of people that I can go back on. >> That is brilliant advice. I love that having a... Isn't that great Tyler? Having a personal- >> Yes Yes! >> Board of directors, especially as we look at cybersecurity and the cybersecurity skills gap Tyler has been, I think it's in its 5th year now, which is there's so much opportunity. What we saw in the threat landscape in the last 18, 19 months during the pandemic was this explosion and the attack surface, ransomware becoming a word that even my mom knows these days. What do you advise Tyler for, you talked about really making people much more aware of all of the opportunities within cyber, but when you think about how you would get women interested in cybersecurity specifically, what are some of the key pieces of advice you would offer? >> Well, again, I think I love the board of directors. I love that. That is brilliant, but I really think that it is about finding mentors, and it is about doing the research, and really asking questions. Because if you reach out to someone on LinkedIn, you know, they may just not respond, but chances are some someone will and, you know, most people in this community are very willing to help. And, you know, I found that to be great. I mean, I've got my board of directors too. I realize that now. (Debbie laughs) But I also like to help other people as well, that are just kind of entering into the field or if they're changing their careers. And it's not necessarily just women, it's people that are interested in getting into an aspect of this industry. And this is a industry where, you know, you can jump from this, to this, to this, to this. I mean, I think that I've had six different major career shifts still within the cybersecurity realm. So, just because you start off doing one thing doesn't mean that that's what you're going to do forever. There're so many different areas. And it's really interesting. I think about my 11 year old niece and she may very well have a job someday, that doesn't even exist right now. That's how quickly cyber and everything connected is moving. And if you think about it, we are connected, there is a cyber component to every single thing that we do, and that's going to continue to expand and continue to grow. And we need more people to be interested, and to want to get into these careers. And I think also it's important for younger girls to let them know these careers are really fun and they're extremely rewarding. And I mean, I hate to use this as an incentive, but there's also a lot of money that can be made too, and that's an incentive to get, you know, women and girls into these careers as well. >> And Tyler, I think you're right. In addition to that, you're always going to have a job. And I think cyber is a great career for someone that are lifelong learners, because like you said, your 11 year old niece, the job, when she graduates from college, she may have, probably doesn't even exist today. And so I think you have to be a lifelong learner. I think one of the things that people may not be aware of is, you know, for women who may have gone the non-traditional route and got degrees later in life, or took time off to raise children and want to come back to work, cyber security is something that, you know, doesn't have to be a nine to five job. I have, it happens to be a gentlemen on my team, who has to get kids on the bus and off the bus. And so we figured out how, you know, he gets up and he works for a couple hours, puts kids on the bus, is in the office. And then he gets the kids off. And once they've had dinner and gone to bed, he puts in a couple more hours. And I think, you know, people need to be aware of, there is some flexibility, there is flexibility in cyber jobs. I mean, it's not a nine to five job, it's not like banking. Well, if you were teller, and your hours are when the bank is open, cyber is 7/24 and jobs can be flexible. And I think people need to be aware of that. >> I agree on the flexibility front, and people also need to be flexible themselves. I do want to ask you both, we're getting low on time, but I've got to ask you, how do you get the confidence, to be, like you said, back in the day, in the room, maybe the only female and I've been in that as well, even in marketing, product marketing years ago. How do you get the confidence to continue moving forward? Even as someone says, "You're only here because you're a female." Tyler, what's your advice to help young women and young men as well fight any sort of challenges that are coming their way? >> I had a mentor when I first moved to the Defense Intelligence Agency, I had an Office Chief and she said to me, "Tyler, you're a Senior Intelligence Officer, you always take a seat at the table. Do not let anyone tell you that you cannot have a seat at the table." And you know, that was good advice. And I think confidence is great. But courage is something that's much more important, because courage is what leads up to confidence. And you really have to believe in yourself and do things that you know are right for you, not because you think it's going to make other people happy. And I think, you know, as women, it's really finding that courage to be brave and to be strong and to be willing to stand out, you know, alone on something, because it's what you care about and what you believe in. And that's really what helps kind of motivate me. >> I love that courage. Debbie, what are your thoughts? >> (laughs) So I was going to say, this is going to be really hard to believe, but when I was 16 years old, I was so shy that if I went to a restaurant and someone served me stone cold food, I wouldn't say a word. I would just eat it. If I bought something in a store and I didn't like it, I'd refuse, I just couldn't bring myself to go to that customer service desk and return it. And my first job in high school, was it a fast food place. And I worked for a gentleman who was a little bit of a tyrant, but you know, I learned how to get a backbone very quickly. And I would have to say now looking back, he was probably my first mentor without even trying to do that. He mentored me on how to believe in myself and how to stand up for what's right. So, Tyler, I completely agree with you. And you know, that's something that people think when they get a mentorship, sometimes it's someone going to mentor them on, you know, something tactical, something they want to know how to do, but sometimes what you need to be mentored in, could be, "How do I believe in myself?" Or "How do I find the courage to be that the only female in the room?" And I think that is where some of that mentorship comes from and, you know, I think, you know, if we go back to mentoring at the middle school, there's lots of opportunities, career fairs, the first robotically, get the middle school level, gives all of us an opportunity to sort of mentor girls at that level. And for all the guys out there who have daughters, this is, you know, how to... It's not like you can get a parenting checklist, "Teach my kid courage." And Tyler, I love that word, but I think that's something that we all need to aspire to bring out in others. >> I love that. I love that. >> Okay with that, I think I love both of your stories, are zig-zaggy in certain ways, one in a more direct cybersecurity path, Debbie with yours. Tyler, yours, very different coming from the music industry. But you both have such great advice. It's really, I would say, I'm going to add that, open your mind to be open to, you can do anything. As Tyler said, there's a very great possibility that right now the job that your niece who's 11 is going to get in the next 10 years, doesn't exist yet. How exciting is that? To have the opportunity to be open-minded enough and flexible enough to say, "I'm going to try that." And I'm going to learn from my mentors, whether it's a fast food cook, which I wouldn't think would be a direct mentor, and recognizing years later, "Wow, what an impact that person had on me, having the courage to do what I have." And so I would ask you like each one more question in terms of just your inspiration for what you're currently doing. Debbie, as the leader of security for NETSCOUT, what inspires you to continue in your current role and seek more? >> So, I'm a lifelong learner. So, I love to learn cybersecurity. You know, every day is a different day. So, it's definitely the ability to continue to learn and to do new things. But the second thing is, is I think I've always been, I don't want to call it a fixer-upper because cybersecurity isn't a fixer-upper, I'm just always wanted to improve upon things. If I've seen something that I think can do better, or a product that could have something new or better in it, you know, that's what excites me is to give people that feedback and to improve on what we've had out there. You know, you had mentioned, we've got this block of jobs that we can't fill. We have to give feedback and how we get the tools and what we have today smarter, so that if there are less of us, we're working smarter and not harder. And so if there is some low-level tasks that we could put back into tools, and talk to vendors and have them do this for us, that's how I think we start to get our way sort of out of the hole. Tyler, any thoughts on that? >> I again, I love that answer. I mean, I think for me, you know, I do like, it's that problem solving thing too. But for me it's also about, it's about compassion. And when I see, you know, a story of some child that's been involved in some kind of cyber bullying attack, or a company that has been broken into, I want to do whatever I can to help people, and to teach people to really protect themselves, so that they feel empowered and they're not afraid of cyber security. So for me, it's also really that drive to really make a difference and really help people. >> And you've both done, I'm sure, so much of that made such a big difference in many communities in which you're involved. I thank you so much for sharing your journeys with me on the program today, and giving such great pointed advice to young men and women, and even some of the older men and women out there that might be kind of struggling about, where do I go next? Your advice is brilliant, ladies. Thank you so much. It's been a pleasure talking with you. >> Thank you. >> Thank you. >> For Debbie Briggs and Tyler Cohen Wood, I'm Lisa Martin. You've been watching this Cube Conversation. (upbeat music)

(upbeat music) >> Hello everyone, welcome back to theCUBE's coverage of splunk.com 2021 virtual. We're here live in the Splunk studios. We're all here gettin all the action, all the stories. Garth Fort, senior vice president, Chief Product Officer at Splunk is here with me. CUBE alumni. Great to see you. Last time I saw you, we were at AWS now here at Splunk. Congratulations on the new role. >> Thank you. Great to see you again. >> Great keynote and great team. Congratulations. >> Thank you. Thank you. It's a lot of fun. >> So let's get into the keynote a little bit on the product. You're the Chief Product Officer. We interviewed Shawn Bice, who's also working with you as well. He's your boss. Talk about the, the next level, cause you're seeing some new enhancements. Let's get to the news first. Talk about the new enhancements. >> Yeah, this was actually a really fun keynote for me. So I think there was a lot of great stuff that came out of the rest of it. But I had the honor to actually showcase a lot of the product innovation, you know, since we did .conf last year, we've actually closed four different acquisitions. We shipped 43 major releases and we've done hundreds of small enhancements, like we're shipping code in the cloud every six weeks and we're shipping new versions twice a year for our Splunk Enterprise customers. And so this was kind of like if you've seen that movie Sophie's Choice, you know, where you have to pick one of your children, like this was a really hard, hard thing to pick. Cause we only had about 25 minutes, but we did like four demos that I think landed really well. The first was what we call ingest actions and you know, there's customers that are using, they start small with gigabytes and they go to terabytes and up to petabytes of data per day. And so they wanted tools that allow them to kind of modify filter and then route data to different sort of parts of their infrastructure. So that was the first demo. We did another demo on our, our visual playbook editor for SOAR, which has improved quite a bit. You know, a lot of the analysts that are in the, in the, in the SOC trying to figure out how to automate responses and reduce sort of time to resolution, like they're not Python experts. And so having a visual playbook editor that lets them drag and drop and sort of with a few simple gestures create complex playbooks was pretty cool. We showed some new capabilities in our APM tool. Last year, we announced we acquired a company called Plumbr, which has expertise in basically like code level analysis and, and we're calling it "Always On" profiling. So we, we did that demo and gosh, we did one more, four, but four total demos. I think, you know, people were really happy to see, you know, the thing that we really tried to do was ground all of our sort of like tech talk and stuff that was like real and today, like this is not some futuristic vision. I mean, Shawn did lay out some, some great visions, visionary kind of pillars. But, what we showed in the keynote was I it's all shipping code. >> I mean, there's plenty of head room in this market when it comes to data as value and data in motion, all these things. But we were talking before you came on camera earlier in the morning about actually how good Splunk product and broad and deep the product portfolio as well. >> Garth: Yeah. >> I mean, it's, I mean, it's not a utility and a tooling, it's a platform with tools and utilities. >> Garth: Yeah >> It's a fully blown out platform. >> Yeah. Yeah. It is a platform and, and, you know, it's, it's one that's quite interesting. I've had the pleasure to meet a couple of big customers and it's kind of amazing, like what they do with Splunk. Like I was meeting with a large telco on the east coast and you know, they actually, for their set top boxes, they actually have to figure out in real time, which ads to display and the only tool they could find to process 15 million events in real time, to decide what ad to display, was Splunk. So that was, that was like really cool to hear. Like we never set out to be like an ad tech kind of platform and yet we're the only tool that operates at that level of scale and that kind of data. >> You know, it's funny, Doug Merritt mentioned this in my interview with him earlier today about, you know, and he wasn't shy about it, which was great. He was like, we're an enabling platform. We don't have to be experts in all these vertical industries >> Garth: Yep >> because AI takes care of that. That's where the machine learning >> Garth: Yeah >> and the applications get built. So others are trying to build fully vertically integrated stacks into these verticals when in reality they don't have to, if they don't want it. >> Yeah, and Splunk's kind of, it's quite interesting when you look across our top 100 customers, you know, Doug talks about like the, you know, 92 of the fortune 100 are kind of using Splunk today, but the diversity across industries and, you know, we have government agencies, we have, you know, you name the retail or the vertical, you know, we've got really big customers, they're using Splunk. And the other thing that I kind of, I was excited about, we announced the last demo I forgot was TruSTAR integration with Enterprise Security. That's pretty cool. We're calling that Splunk Threat Intelligence. And so That was really fun and we only acquired, we closed the acquisition to TruSTAR in May, but the good news is they've been a partner with us like for 18 months before we actually bought em. And so they'd already done a lot of the work to integrate. And so they had a running start in that regard, But other, one other one that was kind of a, it was a small thing. I didn't get to demo it, but we talked about the, the content pack for application performance monitoring. And so, you know, in some ways we compete in the APM level, but in many ways there's a ton of great APM vendors out there that customers are using. But what they wanted us to do was like, hey, if I'm using APM for that one app, I still want to get data out of that and into Splunk because Splunk ends up being like the core repository for observability, security, IT ops, Dev Sec Ops, et cetera. It's kind of like where the truth, the operational truth of how your systems works, lives in Splunk. >> It's so funny. The Splunk business model has actually been replicated. They call it data lake, whatever you want to call it. People are bringing up all these different metaphors. But at the end of the day, if you guys can create a value proposition where you can have data just be, you know, stored and dumped and dumped into whatever they call it stored in a way >> Garth: We call it ingest >> Ingested, ingested. >> Garth: Not dumped. >> Data dump. >> Garth: It's ingested. >> Well, I mean, well you given me a plan, but you don't have to do a lot of work to store just, okay, we can only get to it later, >> Garth: Yep. >> But let the machines take over >> Garth: Yep. >> With the machine learning. I totally get that. Now, as a pro, as a product leader, I have to ask you your, your mindset around optimization. What do you optimize for? Because a lot of times these use cases are emerging. They just pop out of nowhere. It's a net new use case that you want to operationalize. So balancing the headroom >> Yep. >> Or not to foreclose those new opportunities for customers. How are customers deciding what's important to them? How do you, because you're trying to read the tea leaves for the future >> Garth: A little bit, yeah. >> and then go, okay, what do our customers need, but you don't want to foreclose anything. How do you think about product strategy around that? >> There's a ton of opportunity to interact with customers. We have this thing called the Customer Advisory Board. We run, I think, four of them and we run a monthly. And so we got an opportunity to kind of get that anecdotal data and the direct contact. We also have a portal called ideas.splunk.com where customers can come tell us what they want us to build next. And we look at that every month, you know, and there's no way that we could ever build everything that they're asking us to, but we look at that monthly and we use it in sort of our sprint planning to decide where we're going to prioritize engineering resources. And it's just, it's kind of like customers say the darndest things, right? Sometimes they ask us for stuff and we never imagined building it in a million years, >> John: Yeah. >> Like that use case around ads on the set top box, but it's, it's kind of a fun place to be like, we, we just, before this event, we kind of laid out internally what, you know, Shawn and I kind of put together this doc, actually Shawn wrote the bulk of it, but it was about sort of what do we think? Where, where can we take Splunk to the next three to five years? And we talked about these, we referred to them as waves of innovation. Cause you know, like when you think about waves, there's multiple waves that are heading towards the beach >> John: Yeah. >> in parallel, right? It's not like a series of phases that are going to be serialized. It's about making a set of investments. that'll kind of land over time. And, and the first wave is really about, you know, what I would say is sort of, you know, really delivering on the promise of Splunk and some of that's around integration, single sign-on things about like making all of the Splunk Splunk products work together more easily. We've talked a lot in the Q and a about like edge and hybrid. And that's really where our customers are. If you watch the Koby Avital's sort of customer keynote, you know, Walmart by necessity, given their geographic breadth and the customers they serve has to have their own infrastructure. They use Google, they use Azure and they have this abstraction layer that Koby's team has built on top. And they use Splunk to manage kind of, operate basically all of their infrastructure across those three clouds. So that's the hybrid edge scenario. We were thinking a lot about, you mentioned data lakes. You know, if you go back to 2002, when Splunk was founded, you know, the thing we were trying to do is help people make sense of log files. But now if you talk to customers that are moving to cloud, everybody's building a data lake and there's like billions of objects flowing into millions of these S3 buckets all over the place. And we're kind of trying to think about, hey, is there an opportunity for us to point our indexing and analytics capability against structured and unstructured data and those data lakes. So that that'll be something we're going to >> Yeah. >> at least start prototyping pretty soon. And then lastly, machine learning, you know, I'd say, you know, to use a baseball metaphor, like in terms of like how we apply machine learning, we're like in the bottom of the second inning, >> Yeah. >> you know, we've been doing it for a number of years, but there's so much more. >> There's so, I mean, machine learning is only as good as the data you put into the machine learning. >> Exactly, exactly. >> And so if you have, if you have gap in the data, the machine learning is going to have gaps in it. >> Yeah. And we have, we announced a feature today called auto detect. And I won't go into the gory details, but effectively what it does is it runs a real-time analytics job over whatever metrics you want to look at and you can do what I would consider more statistics versus machine learning. You can say, hey, if in a 10 minute period, like, you know, we see more errors than we see on average over the last week, throw an alert so I can go investigate and take a look. Imagine if you didn't have to figure out what the right thresholds were, if we could just watch those metrics for you and automatically understand the seasonality, the timing, is it a weekly thing? Is it a monthly thing? And then like tell you like use machine learning to do the anomaly detection, but do it in a way that's more intelligent than just the static threshold. >> Yeah. >> And so I think you'll see things like auto detect, which we announced this week will evolve to take advantage of machine learning kind of under the covers, if you will. >> Yeah. It was interesting with cloud scale and the data velocity, automations become super important. >> Oh yeah. >> You don't have a lot of new disciplines emerge, like explainable AI is hot right now. So you got, the puck is coming. You can see where the puck is going. >> Yeah >> And that is automation at the app edge or the application layer where the data has got to be free-flowing or addressable. >> Garth: Yeah. >> This is something that is being talked about. And we talked about data divide with, with Chris earlier about the policy side of things. And now data is part of everything. It's part of the apps. >> Garth: Yeah. >> It's not just stored stuff. So it's always in flight. It should be addressable. This is what people want. What do you think about all of that? >> No, I think it's great. I actually just can I, I'll quote from Steve Schmidt in, in sort of the keynote, he said, look like security at the end of the day is a human problem, but it kind of manifests itself through data. And so being able to understand what's happening in the data will tell you, like, is there a bad actor, like wreaking havoc inside of my systems? And like, you can use that, the data trail if you will, of the bad actor to chase them down and sort of isolate em. >> The digital footprints, if you will, looking at a trail. >> Yeah. >> All right, what's the coolest thing that you like right now, when you look at the treasure trove of, of a value, as you look at it, and this is a range of value, Splunk, Splunk has had customers come in with, with the early product, but they keep the customers and they always do new things and they operationalize it >> Garth: Yep. >> and another new thing comes, they operationalize it. What's the next new thing that's coming, that's the next big thing. >> Dude that is like asking me which one of my daughters do I love the most, like that is so unfair. (laughing) I'm not going to answer that one. Next question please. >> Okay. All right. Okay. What's your goals for the next year or two? >> Yeah, so I just kind of finished roughly my first 100 days and it's been great to, you know, I had a whole plan, 30, 60, 90, and I had a bunch of stuff I wanted to do. Like I'm really hoping, sort of, we get past this current kind of COVID scare and we get to back to normal. Cause I'm really looking forward to getting back on the road and sort of meeting with customers, you know, you can meet over Zoom and that's great, but what I've learned over time, you know, I used to go, I'd fly to Wichita, Kansas and actually go sit down with the operators like at their desk and watch how they use my tools. And that actually teaches you. Like you, you come up with things when you see, you know, your product in the hands of your customer, that you don't get from like a CAB meeting or from a Zoom call, you know? >> John: Yeah, yeah. >> And so being able to visit customers where they live, where they work and kind of like understand what we can do to make their lives better. Like that's going to, I'm actually really excited to gettin back to travel. >> If you could give advice to CTO, CISO, or CIO or a practitioner out there who are, who is who's sitting at their virtual desk or their physical desk thinking, okay, the pandemic, were coming through the pandemic. I want to come out with a growth strategy, with a plan that's going to be expansive, not restrictive. The pandemic has shown what's what works, what doesn't work. >> Garth: Sure. >> So it's going to be some projects that might not get renewed, but there's doubling down on, certainly with cloud scale. What would advice would you give that person when they start thinking about, okay, I got to get my architecture right. >> Yeah. >> I got to get my playbooks in place. I got to get my people aligned. >> Yeah >> What's what do you see as a best practice for kind of the mindset to actual implementation of data, managing the data? >> Yeah, and again, I'm, I'm, this is not an original Garth thought. It actually came from one of our customers. You know, the, I think we all, like you think back to March and April of 2020 as this thing was really getting real. Everybody moved as fast as they could to either scale up or scale scaled on operations. If you were in travel and hospitality, you know, that was, you know, you had to figure how to scale down quickly and like what you could shut down safely. If you were like in the food delivery business, you had to figure out how you could scale up, like Chipotle hit two, what is it? $2 billion run rate on delivery last year. And so people scrambled as fast as they could to sort of adapt to this new world. And I think we're all coming to the realization that as we sort of exit and get back to some sense of new normal, there's a lot of what we're doing today that's going to persist. Like, I think we're going to have like flexible rules. I don't think everybody's going to want to come back into the office. And so I think, I think the thing to do is you think about returning to whatever this new normal looks like is like, what did we learn that was good. And like the pandemic had a silver lining for folks in many ways. And it sucked for a lot. I'm not saying it was a good thing, but you know, there were things that we did to adapt that I think actually made like the workplace, like stronger and better. And, and sort of. >> It showed that data's important, internet is important. Didn't break, the internet didn't break. >> Garth: Correct. >> Zoom was amazing. And the teleconferencing with other tools. >> But that's kind of, just to sort of like, what did you learn over the last 18 months that you're going to take for it into the next 18 years? You know what I mean? Cause there was a lot of good and I think people were creative and they figured out like how to adapt super quickly and take the best of the pandemic and turn it into like a better place to work. >> Hybrid, hybrid events, hybrid workforce, hybrid workflows. What's what's your vision on Splunk as a tier one enterprise? Because a lot of the news that I'm seeing that's, that's the tell sign to me in terms of this next growth wave is big SI deals, Accenture and others are yours working with and you still got the other Partnerverse going. You have the ecosystems emerging. >> Garth: Yep. >> That's a good, that means your product's enabling people to make money. >> Garth: Yeah. Yeah, yeah, yeah. >> And that's a good thing. >> Yeah, BlueVoyant was a great example in the keynote yesterday and they, you know, they've really, they've kind of figured out how, you know, most of their customers, they serve customers in heavily regulated industries kind of, and you know, those customers actually want their data in a Splunk tenant that they own and control and they want to have that secure boundary around that. But BlueVoyant's figured out how they can come in and say, hey, I'm going to take care of the heavy lifting of the day-to-day operations, the monitoring of that environment with the security. So, so BlueVoyant has done a great job sort of pivoting and figuring out how they can add value to customers and do, you know, because they they're managing not just one Splunk instance, but they're managing 100s of Splunk cloud instances. And so they've got best practices and automation that they can play across their entire client base. And I think you're going to see a lot more of that. And, and Teresa's just, Teresa is just, she loves Partners, absolutely loves Partners. And that was just obvious. You could, you could hear it in her voice. You could see it in her body language, you know, when she talked about Partnerverse. So I think you'll see us start to really get a lot more serious. Cause as big as Splunk is like our pro serve and support teams are not going to scale for the next 10,000, 100,000 Splunk customers. And we really need to like really think about how we use Partners. >> There's a real growth wave. And I, and I love the multiples wave in parallel because I think that's what everyone's consensus on. So I have to ask you as a final question, what's your takeaway? Obviously, there's been a virtual studio here where all the Splunk executives and, and, and customers and partners are here. TheCUBE's here doing all the presentations, live by the way. It was awesome. What would you say the takeaway is for this .conf, for the people watching and consuming all the content online? A lot of asynchronous consumption would be happening. >> Sure. >> What's your takeaway from this year's Splunk .conf? >> You know, I, it's hard cause you know, you get so close to it and we've rehearsed this thing so many times, you know, the feedback that I got and if you look at Twitter and you look at my Slack and everything else, like this felt like a conf that was like kind of like a really genuine, almost like a Splunk two dot O. But it's sort of true to the roots of what Splunk was true to the product reality. I mean, you know, I was really careful with my team and to avoid any whiff of vaporware, like what were, what we wanted to show was like, look, this is Splunk, we're acquiring companies, you know, 43 major releases, you know, 100s of small ones. Like we're continuing to innovate on your behalf as fast as we can. And hopefully this is the last virtual conf. But even when we go back, like there was so much good about the way we did this this week, that, you know, when we, when we broke yesterday on the keynote and we were sitting around with the crew and it kind of looking at that stage and everything, we were like, wow, there is a lot of this that we want to bring to an in-person event as well. Cause so for those that want to travel and come sit in the room with us, we're super excited to do that as soon as we can. But, but then, you know, there may be 25, 50, 100,000 that don't want to travel, but can access us via this virtual event. >> It's like a time. It's a moment in time that becomes a timeless moment. That could be, >> Wow, did you make that up right now? >> that could be an NFT. >> Yeah >> We can make a global cryptocurrency. Garth, great to see you. Of course I made it up right then. So, great to see you. >> Air bump, air bump? Okay, good. >> Okay. Garth Fort, senior vice president, Chief Product Officer. In theCUBE here, we're live on site at Splunk Studio for the .conf virtual event. I'm John Furrier. Thanks for watching. >> All right. Thank you guys. (upbeat music)