(upbeat music) >> Hello, and welcome back to theCUBE's coverage of Red Hat Summit '21 virtual conference. I'm John Furrier, host of theCUBE. We are here with Kevin Martelli, Principal Software Engineer at KPMG, joining the conversation. Kevin, great to see you. Thanks for coming on. >> John, thanks a lot for having me. >> So obviously Red Hat, a lot of action, cloud native, part of IBM now. A lot of talk going on around this growth around cloud. Massive new opportunities, new modern applications being shaped in, super exciting opportunities. So first, before we get into all that, tell us about your role at KPMG. >> Sure John, thanks. So my role at KPMG, I'm one of our cloud leaders at KPMG where I really help both from an internal perspective, so helping our internal enablement and digitalization, as well as, more importantly, helping to deliver solutions and applications to our clients as they go through these digital journeys. And really focusing on containerization and enabling it through the cloud. >> John: You guys have done a lot of AI work, I know which is cutting edge, it's pretty much data-driven. I mean, AI is, what everyone talks about, but underlying AI is automation, data, machine learning, really dealing with kind of new types of datasets, not just dealing with existing structures. You have a new platform called Ignite. Tell us what that is. What do you guys solve? What was the problem statement? And what's going on with it? >> Yeah, John, thanks a lot for asking. So Ignite, it's something that we developed internally initially, and it really helped to solve our AI initiatives. We called it our AI platform, but it's moreso an ecosystem. And it solves not only our own internal needs and internal use cases, but also choose to help support and deliver these solutions to the clients. One of the foundational principles of our platform is it's built on top of containerization, which we know is a hot area now, today in the marketplace really gives you the ability for scalability, flexibility, security, et cetera, but more important, what we're seeing large scale of adoptions in our clients, is using this platform to really get value out of both unstructured and structured data in a way that they're able to do this in a secured fashion and then easily get it deployed. It's a pretty scalable platform, and something that we've just recently received the patent for it. >> So what was the internal conversation to put this together? Was it the fact that there was business needs? Cloud native gave you that scale advantage? What was some of the drivers behind Ignite? 'Cause this is like, was it IoT? Was it, take us through the mindset. What were some of the first principles around building this? >> Hey John nice, it's a good question. And actually to be fair, this was probably a little bit before the time of IoT and some of these newer technologies were coming up. At this time, we were really kind of scratching on the surface of data science and advanced analytics. And what really generated the need for this is as you could imagine, working in a consultancy firm and many of our clients deal with tons of contracts and the lightboard documents for financial services, there was much rich information, these unstructured data documents and we had no way to get this information out. So really it was generated out of the need to get information out of a lot of these contractual documents that we had and pinpoint specific information. So really taking it holistically on ingestion on transformations, running NLP, algorithms, it really evolved into a whole end to end complete platform, running on top of a containerized ecosystem, such as OpenShift. >> John: Yeah, I think just not to go on a tangent here but I think one of the conversations we've been having on all these events and certainly with COVID was the highlight of all these silos. And you know the old days was about break down the silos. But now with containers and cloud scale, you can extract out data, kind of create that horizontal data plane if you will, or view observation space, some call it. This just seems to be a huge trend you guys were on it early. How has that, what's your take on that? The silos used to be kind of like an advantage if you had a monolithic application but now you have a lot of diverse distributed databases. What's your take? >> Kevin: Yeah, it's, it's good. And how we are kind of coining. It is really through the power of, some of the toying in OpenShift that really gives organizations the ability to defer risk. In the sense that allows you to run certain types of workloads on-prem in a private cloud containerized way. It allows you to burst certain other types of workloads into the different CSP provider. So you can get advantage of their scale, their capacity without maybe moving some sensitive data and then another benefit is with some of that vendor lock-in it sometimes clients are concerned about is being able to kind of easily deploy your workloads and applications from one cloud provider to another. And I think as we look at this distributed processing no one client will totally be in one cloud provider. So having the ability to move workloads quickly and fastly where they make sense, where the security and risk is aligned is something that would what makes a successful use cases deployments. >> John: Just let me ask you another question. You guys, KPMG obviously have your own big data effort going on with analytics. You've got clients that you serve and ultimately they have customers as well. So you have that Red Hat equation. What are some of the advantages that you guys see as your firm and your clients with Red Hat analytics, 'cause this becomes ultimately the number one conversation. Like, okay, what's in it for me? >> Yeah. That's a good point. I would say we're seeing a few things. Some of them are highlighted. One is, as you're well aware, we chose Red Hat's OpenShift as one of our strategic options to deploy our platform. And whenever you're deploying these platforms it's very important that you have the flexibility the agility, and the ability to scale and Red Hat underneath the hood really helps take care of a lot of that, for you in a way that not only can you do it on your own as mentioned earlier, your private cloud but also onto the public CSPs and multiple CSPs. In addition, some of the other things I think that we saw that were very beneficial, a lot of times as an application user. So application users of ignite, the developers, the data scientists, the business users, the analysts, they all need to interact with the platform. They want to worry about getting the insights about getting the efficiencies in the platform. They don't want to worry about how the infrastructure's being put together, how the workloads are being moved how the scalability is occurring, et cetera and Red Hat really takes a lot of that away from you having to worry about it. And one of the other things that's also important is, is we have a strategic relationship with Red Hat. And as we look to help to enhance and develop these capabilities and experiences as our clients are doing private cloud, hybrid cloud and multi-cloud, we're really going to be able to let them take the power of open source, into their own control and how they want to deploy it in themselves. >> Well, got you on the topic there. I got to ask you the question. What would you say to the people out there that haven't really kicked the tires on Red Hat in a while? What's the modern update? How would you describe the current situation at Red Hat for people who are going to re-look and or bring the Red Hat conversation up a notch? >> Yeah, it's a good question. I think we see this in any type of software in the industry today. There's so many choices and there's so many options out there. And how do you choose the right source for the right use case? For the right client, for the right company? And how we always like to talk with clients is that yes, there are a lot of choices in there and the orchestration for the standardization but when you're looking for something that's celebrated in the market that has the security built into it that many organizations are looking for that gives you the flexibility without having to do a lot of additional operational overhead of moving from on-prem into the cloud and the way that it can scale and kind of make the overall ecosystem operations and deployments easier, it's one of the benefits that we see have gone with a tool like Red Hat OpenShift. >> Well, Kevin, I really appreciate the comments there and on Red Hat, that's awesome. Red Hat Summit, honestly, a big event around Red Hat and future cloud and modern applications. So I got to ask you as a software engineering leader in the industry, you got to be pretty excited about artificial intelligence and machine learning as it relates to, what it can be doing for changing the software development paradigm. Obviously there's also the no code, low code, serverless. You've got cloud native, you've got containers you got all this new capability. So how does, how do you see those trends? What are the big trends around machine learning and AI as it relates to someone who's going to be building modern applications in the cloud. Because certainly there's a huge ups upside there. Some are saying that if you don't have AI that's going to be a table stakes and we'll lower the valuation of the software or the application. What's your take on all these big trends around AI? >> Yeah, I agree with that. We've actually done several studies. And what we're hearing industry leaders saying is it was quite a few things. One is, we at KPMG, COVID-19 whiplash. And really what that means is that the pace and acceleration of adoption in AI has been tremendous over the COVID 19 period of our pandemic period. And so much so that industry leaders are a little bit concerned about how fast this adoption is going. And is it going too fast? In addition, we recently published a study called Thriving In An AI world where we were able to identify that business leaders and insiders are really bullish on to your point of using AI and ML to make some poor, critical decisions. How can we make vaccines? What's the distribution process? Fraudulent analytics where financial services. However, what I will say is we're still seeing a lot, a lot of questions and challenges around AI. Its security, its ethics associated to it. How you keep managing governing your process then privacy associated to it. So there's a lot of points around those areas. I think that industries are still trying to struggle and figure out how to solve for. And one of the things that we are hearing is that what the new administration there's different think tanks and industry leaders that are feeling that the new administration, while open to a lot of these advanced techniques and technologies are going to put a little bit more rigor around and regulations around how AI can be used in the marketplace. So hopefully that would give some companies guidance around these security and privacy and ethics concerns. >> Yeah, it's interesting. I was talking to a friend the other day who's a leader at a big company that's a customer of Red Hat and a lot of other clouds as well. And we were joking about the agility speed, oh, agility and speed. Of course, yeah, you get that with here but you got a lot of fast and loose situations going here. You got to know when to put the pedal to the metal. When there's a straight narrow, we can really kind of gas it with AI and machine learning and then know where the potential curves are. See, will use that metaphor because you can go fast but with speed comes dangerous new things for breakage. Is always, and you're seeing that all the time. You're seeing that, with software because you can push new update, but still, when you talk about operational integrity and security fast and loose, isn't always the best way to go. But if you know there's a straight and narrow, you can really push it. This was what we were saying, he's like, "Hey, we know when to go straight and narrow and go fast. And then when to slow it down, pull it back." What's your take on that? What's your assessment? >> No, I agree. I think you hit some valid points there. And sometimes what we do is we take some antiquated processes and we overlay them into these newer technologies and we try to think them as being the same way and they may not always hold true. But it's not only kind of the fast and narrow and then putting things in that maybe a little bit more simplistic, but it's also there's a whole change around how you productionalized. How do you get these things into deployment? How do you monitor these over time? So some of those biases or some of those privacy concerns don't end up creeping up into the algorithm over time. I still think that will work here and from industries. There are still struggles around that. There's still struggles around. There's a lot of technologies that can do a lot of these same things. Our business processes don't always align. And then how do we really take something from an innovation from a POC into production? Is there a fast track for something that is straightened narrow and something that has a little bit more complexity? But what we're seeing today, there's a lot sort of spout at the same road, which makes bringing more complex AI algorithms into production. Challenging. >> Yeah. And there's always that big trend of day two operations. Which is, hey, you deploy it's great. And then, okay, wait a minute stop, set in a break. We need better monitoring. We need better data analytics. What's instrumented. What's not. What services are being generated and terminated. These are all big cloud native kind of themes. With that, I got to ask you from a customer standpoint, these are new first-generation problems at scale that with this new cloud native environment, the pros and cons. How do you guys talk to customers? What are some of the things you're seeing around the challenges that they face with analytics? All these analytic activity? >> Kevin: Yeah. So I think one of the challenges and we've probably heard this year in year out is around data literacy. Like really having our folks understand the data and empowering them to be successful in the organization. And to be fair I would say data literacy was a little bit more narrowly focused in an organizations who needed it. I need some analysts to use it. I needed some data scientists and engineers, but what we're starting to see now is there's larger programs across the board where it's more holistic at an organizational level. Everyone should be involved in data. Everyone should be able to do their own reporting. So really data literacy and getting data kind of into the arms of the folks is important. Some of the other ones that we've also kind of talked to about it, and they kind of go hand in hand and maybe a little bit on our prior conversation was the technologies. Technology especially in open source is exploding. And as well as commercial. So how do you choose the right technologies the right tools? You don't have too many tools in your toolbox per se but use the ones that are really differentiating and try to standardize on the ones that are more standard. Finally it's bringing those processes and that wrapping them back into the technologies. Again, a little point we hit on earlier but what we're finding is as technology is rapidly increasing, you're able to use it for your analytics. Your processes are still antiquated and legacy processes which makes it a little bit harder for you to really take advantage of what you're trying to achieve in your organization from a digital transformation. And then one final one I would add in there is around the risk that organizations have. So there's a lot of concern about reputational risk. If they're doing these types of activities that people don't understand, the data they don't understand the algorithms. Are there some impacts that can be heard? And they're figuring out how to control that and then how not to. And then I think finally the workforce is, as we know, it's getting the workforce up to speed, retooling where need be and putting their people in the right place to be successful. >> Kevin that's great insight. Thanks so much for coming on theCUBE. I got to ask you one final question. >> Go ahead. One more thing, you mentioned COVID whiplash means a lot of post COVID activity discussions going on. If you look at what's happened with COVID there's been an exposure of all the projects that need to be doubled down on, ones that may not be continuing. People working at home. Honestly, a change of the environment, you mentioned workforce is among others. What do you think the biggest conversation around your customer base or within KPMG right now around some of these growth strategies around post COVID? What are companies thinking around how to deploy the people, process and technology is a big part of this conversation. What is the post COVID general theme that you're seeing among large enterprises and businesses in general? >> I mean, that's a good question. So I think in general, we're seeing the acceleration of digital agendas that may have been pushed out for five years school moving closer. But one of the most interesting things I think that I've gathered out of working with the clients that we're working with is that before to get stuff into production, AI solutions even in any type of smaller production system that was taking months months, several months to get something in production. And it seemed to be once the COVID pandemic hit, organizations can accelerate that journey of the deployment of applications into production in very, very quick timeframes without hindering or impacting any types of control frameworks they have in place, but just working quicker. So I think some of the things I see as we move forward is that these digital channels are going to be push forward more quicker. The data list on POC is good our pilot's good, is long past. It's now they want to see the results in the outputs in the enterprise, in production. And I think they realize that they have the tools to do this in a period of time that is weeks versus months, and in some cases, years. >> So, would you agree then, just as a quick followup to that that obviously when we get back to real life, post COVID that the visibility and the economics and the productivity gains from this new environment is going to stay around longer and probably be permanent. What's your, do you agree with that statement? >> I hope it is. but we are creatures of habit. And sometimes you go back to back to the way that we had done things, but I'm hopeful that they were able to see to be successful in these types of environments and make these types of decisions that those processes that are evolving to take into consideration what we learned. One is terrible pandemic, and be able to apply that to the post pandemic. >> Yeah. Who would have known the word hybrid cloud actually means something more than just cloud technologies? Hybrid events, hybrid workforces, the word hybrid has been kicked around. Kevin, thanks so much for coming on theCUBE for Red Hat Summit coverage. Thanks for coming on. Great insight. >> Thank you, have a great day. >> Thanks. I'm John Farrow with theCUBE here for Red Hat Summit coverage, 2021 virtual. Thanks for watching. (upbeat music)