>> Narrator: Live from Stanford University in Palo Alto, California, it's the CUBE! Covering, Women in Data Science Conference 2018, brought to you by Stanford. >> Welcome back to the CUBE, we are live at Stanford University, we've been here all day at the third annual Women in Data Science Conference, WiDS 2018. This event is remarkable in its growth in scale, in its third year, and that is, in part by the partners and the sponsors that they have been able to glean quite early on. I'm excited to be joined by Vijay Raghavendra, the senior vice president of Merchant Technology and stores as well, from Walmart Labs. Vijay, welcome to the CUBE! >> Thank you, thank you for having me. >> Walmart Labs has been paramount to the success of WiDS, we had Margot Gerritsen on earlier, and I said, "How did you get the likes of a Walmart Labs as a partner?" And, she was telling me that, the coffee-- the coffee shop conversation >> Yeah, the Coupa Cafe! >> That she had with Walmart Labs a few years ago, and said, "Really, partners and sponsors like Walmart have been instrumental in the growth and the scale, of this event." And, we've got the buzz around, so we can hear the people here, but this is the big event at Stanford. There's 177 regional events, 177! In 53 countries. It's incredible. Incredible, the reach. So, tell me a little bit about the... From Walmart Labs perspective, the partnership with WiDS, what is it that really kind of was an "Aha! We've got to do this"? >> Yeah, it's just incredible, seeing all of these women and women data scientists here. It all started with Esteban Arcaute, who used to lead data science at Walmart Labs, and Search, before he moved on to Facebook with Margot. And, Karen in the cafe in Palo Alto, in 2015, I think. And Esteban and I had been talking about how we really expand the leverage of data and data science within Walmart, but more specifically, how we get more women into data science. And, that was really the genesis of that, and, it was really-- credit goes to Esteban, Margot, and Karen for, really, thinking through it, bringing it together, and, here we are. >> Right, I mean bringing it together from that concept, that conversation here at Stanford Cafe to the first event was six months. >> Yeah, from June to November, and, it's just incredible the way they put it together. And, from a Walmart Labs perspective, we were thrilled to be a huge part of it. And, all the way up the leadership chain there was complete support, including my boss Jeremy King, who was all in, and, that really helped. >> Margot was, when we were chatting earlier, she was saying, "It's still sort of surprising," and she said she's been, I think in, in the industry for, 30-plus years, and she said that, she always thought, back in the day, that by the time she was older, this problem would be solved, this gender gap. And she says, "Actually, it's not like it's still stagnant," we're almost behind, in a sense. When I look at the ... women that are here, in Stanford, and those that are participating via those regional events, the livestream that WiDS is doing, as well as their Facebook livestream. You know, the lofty goal and opportunity to reach 100,000 people shows you that there's clearly a demand, there's a need for this. I'd love to get your perspective on data science at Walmart Labs. Tell me a little bit about the team that you're leading, you lead a team of engineers, data scientists, product managers, you guys are driving some of the core capabilities that drive global e-commerce for Walmart. Tell me about, what you see as important for that female perspective, to help influence, not only what Walmart Labs is doing, but technology and industry in general. >> Yeah. So, the team I lead is called Merchant Technology, and my teams are responsible for, almost every aspect of what drives merchandising within Walmart, both on e-commerce and stores. So, within the purview of my teams are everything from the products our customers want, the products we should be carrying either in stores or online, to, the product catalog, to search, to the way the products are actually displayed within a store, to the way we do pricing. All of these are aspects of what my teams are driving. And, data and data science really put me at every single aspect of this. And the reason why we are so excited about women in data science and why getting that perspective is so important, is, we are in the retail business, and our customers are really span the entire spectrum, from, obviously a lot of women shop at Walmart, lot of moms, lot of millennials, and, across the entire spectrum. And, our workforce needs to reflect our customers. That's when you build great products. That's when you build products that you can relate to as a customer, and, to us that is a big part of what is driving, not just the interest in data science, but, really ensuring that we have as diverse and as inclusive a community within Walmart, so we can build products that customers can really relate to. >> Speaking of being relatable, I think that is a key thing here that, a theme that we're hearing from the guests that we're talking to, as well as some of the other conversations is, wanting to inspire the next generation, and helping them understand how data science relates to, every industry. It's very horizontal, but it also, like a tech company, or any company these days is a tech company, really, can transform to a digital business, to compete, to become more profitable. It opens up new business models, right, new opportunities for that. So does data science open up so many, almost infinite opportunities and possibilities on the career front. So that's one of the things that we're hearing, is being able to relate that to the next generation to understand, they don't have to fit in the box. As a data scientist, it sounds like from your team, is quite interdisciplinary, and collaborative. >> And, to us that is really the essence of, or the magic of, how you build great products. For us data science is not a function that is sitting on the side. For us, it is the way we operate as we have engineers, product managers, folks from the business teams, with our data scientists, really working together and collaborating every single day, to build great products. And that's, really how we see this evolving, it's not as a separate function, but, as a function that is really integrated into every single aspect of what we do. >> Right. One of the things that we talked about is, that's thematic for WiDS, is being able to inspire and educate data scientists worldwide, and obviously with the focus of helping females. But it's not just the younger generation. Some of the things that we're also hearing today at WiDS 2018 is, there's also an opportunity within this community to reinvigorate the women that have been in, in STEM and academia and industry for quite a while. Tell me a little bit more about your team and, maybe some of the more veterans and, how do you kind of get that spirit of collaboration so that those that, maybe, have been in, in the industry for a while get inspired and, maybe get that fire relit underneath them. >> That's a great question, because we, on our teams, when you look across all the different teams across different locations, we have a great mix of folks that bring very different, diverse experiences to the table. And, what we've found, especially with the way we are leveraging data, and, how that is invigorating the way we are... How people come to the table, is really almost seeing the art of what is possible. We are able to have, with data, with data science, we are able to do things that, are, really step functions in terms of the speed at which we can do things. Or, the- for example, take something as simple as search, product search, which is one of the, capabilities we own, or my team is responsible for, but, you could build the machine learning ranking, and, relevance and ranking algorithms, but, when you combine it with, for example, a merchant that really fundamentally understands their category, and you combine data science with that, you can accelerate the learning in ways that is not possible. And when folks see that, and see that in operation that really opens up a whole, slew of other ideas and possibilities that they think about. >> And, I couldn't agree more. Looking at sort of the skillset, we talk a lot about, the obvious technical skillset, that a data scientist needs to have, but there's also, the skills of, empathy, of communication, of collaboration. Tell me about your thoughts on, what is an ideal mix, of skills that that data scientist, in this interdisciplinary function, should have. >> Yeah, in fact, I was talking with a few folks over lunch about just this question! To me, some of the technical skills, the grounding in math and analytics, are table stakes. Beyond that, what we look for in data scientists really starts with curiosity. Are they really curious about the problems they're trying to solve? Do they have tenacity? Do they settle for the more obvious answers, or do they really dig into, the root cause, or the root, core of the problems? Do they have the empathy for our customers and for our business partners, because unless you're able to put yourself in those shoes, you're going to be approaching at, maybe, in somewhat of an antiseptic way? And it doesn't really work. And the last, but one of the most important parts is, we look for folks who have a good sense for product and business. Are they able to really get into it, and learn the domain? So for example, if someone's working on pricing, do they really understand pricing, or can they really understand pricing? We don't expect them to know pricing when they come in, but, the aptitude and the attitude is really, really critical, almost as much as the core technical skills, because, in some ways, you can teach the technical skills, but not some of these other skills. >> Right, and that's an interesting point that you bring up, is, what's teachable, and, I won't say what's not, but what might be, maybe not so natural for somebody. One of the things, too, that is happening at WiDS 2018 is the first annual Datathon. And, Margot was sharing this huge number of participants that they had and they set a few ground rules like wanting the teams to be 50% female, but, tell us about the Datathon from your global visionary sponsorship level; what excites you about that in terms of, the participation in the community and the potential of, "Wow, what's next"? >> Yeah... So, it's hugely exciting for us, just seeing the energy that we've seen. And, the way people are approaching different problems, using data to solve very different kinds of problems ... across the spectrum. And for us, that is a big part of what we look for. For us it is really about, not just coming up with a solution, that's in search of a problem, but really looking at real-world problems and looking at it from the perspective of, "Can I bring data, can I bring data science to bear on this problem?", to solve it in ways that, either are not possible, or can accelerate the way we would solve the problems otherwise. And that is a big part of what is exciting. >> Yeah, and the fact that the impact that data science can make to, every element of our lives is, like I said before, it's infinite, the possibilities are infinite. But that impact is something that, I think, how exciting to be able to be in an industry or a field, that is so pervasive and so horizontal, that you can make a really big social impact. One of they other things, too, that Margot said. She mentioned that the Datathon should be fun, and I loved that, and also have an element of creativity. What's that balance of, creativity in data science? Like, what's the mixture, because we can be maybe over-creative, and maybe interpret something that's in a biased way. What is your recommendation on how much creativity can creep into, and influence, positively, data science? >> Yeah, that's a great question, and there's no perfect answer for it. Ultimately, at least my biases towards using data and data science to, solve real problems. And... As opposed to, pure research, so our focus very much is on applied learning, and applied science. And, to me, within that, I do want the data science to be creative, data scientists to be creative, because, by putting too many guardrails, you limit the way in which they would explore the data, that they may come up with insights that, well, we might not see otherwise. And, which is why, I go back to the point I made, when you have data scientists who fundamentally understand a business, and the business problems we are trying to solve, or the business domains, I think they can then come up with very interesting, innovative ways of looking at the data, and the problem, that you might not otherwise. So, I would by no means want to limit their creativity, but I do have a bias towards ensuring that it is focused on problems we are trying to solve. >> Excellent. Well, Vijay, thank you so much for stopping by the CUBE, congratulations on the continued success of the partnership with WiDS and, we're looking forward to seeing what happens the rest of the year, and we'll probably see you next year at WiDS 2019! >> Absolutely, thank you! >> Excellent, we want to thank you, you're watching the CUBE, live from Stanford University, the third annual Women in Data Science Conference. I am Lisa Martin, I'll be right back after a short break with my next guest. (cool techno music)

>> Narrator: Alumni. (upbeat music) >> Announcer: Live from Stanford University in Palo Alto, California, it's theCUBE. Covering Women in Data Science Conference 2018. Brought to you by Stanford. >> Welcome back to theCUBE, we are live at Stanford University for the third annual Women in Data Science Conference, WiDS. I'm Lisa Martin, very honored to be joined by one of the co-founders of this incredible WiDS movement and phenomenon, Dr. Margot Gerritsen. Welcome to theCUBE! >> It's great to be here, thanks so much for being at our conference. >> Oh, likewise. You were the senior associate dean and director of the Institute for Computational Mathematics and Engineering at Stanford. >> Gerritsen: That's right, yep. >> Wow, that's a mouthful and I'm glad I could actually pronounce that. So you have been, well, I would love to give our audience a sense of the history of WiDS, which is very short. You've been on this incredible growth and scale trajectory. But you've been in this field of computational science for what, 30, over 30 years? >> Yeah, probably since I was 16, so that was 35 years ago. >> Yeah, and you were used to being one of few, or if not the only woman >> That's right. >> In a meeting, in a room. You were okay with that but you realized, you know what? There are probably women who are not comfortable with this and it's probably going to be a barrier. Tell us about the conception of WiDS that you and your co-founders had. >> So, May, 2015, Esteban from Walmart Labs, now at Facebook, and Karen Matthys, who's still very active, you know, one of the organizers of the conference, and I were having coffee at a cafe in Stanford and we were lamenting the fact that at another data science conference that we had been to had only had male speakers. And so we connected with the organizers and asked them why? Did you notice? Because very often people are not even aware, it's just such the norm to only have male speakers, >> Right, right. >> That people don't even notice. And so we asked why is that? And they said, "Well, you know we really tried to find "speakers but we couldn't find any." And that really was, for me, the last straw. I've been in so many of these situations and I thought, you know, we're going to show them. So we joke sometimes, a little bit, we say it's sort of a revenge conference. (laughs) We said, let's show them we can get some really outstanding women, and in fact only women. And that's how it started. Now we were sitting at this coffee shop and I said, "Let's do a conference." And they said, "Well, that would be great, next year." And I said, "No, this year. "Let's just do it. "Let's do it in November." We had six months to put it together. It was just a local conference here. We got outstanding speakers, which were really great. Mostly from the area. And then we started live-streaming because we thought it would be fun to do. And to our big surprise, we had 6,000 people on the livestream just without really advertising. That made us realize, in November 2015, my goodness, we're onto something. And we had such amazing responses. We wanted to then scale up the conference and then you can hire a fantastic conference center in San Francisco and get 10,000 people in like they do, for example, at Grace Hopper. But we thought, why not use online technology and scale it up virtually and make this a global event using the livestream, that we will then provide to people, and asking for regional events, local events to be set up all around the world. And we created this ambassador program, that is now in its second year. the first year the responses were actually overwhelming to us already then. We got 75 ambassadors who set up 75 events around the world >> In about 40 countries. >> This was last year, 2017? >> Yeah, almost exactly 13 months ago, and then this year now we have over 200 ambassadors. We have 177 events in 155 cities in 53 countries. >> That's incredible. >> So we're on every continent apart from Antarctica but we're working on that one. >> Martin: I was going to say, that's probably next year. >> Yeah, that's right. >> The scale, though, that you've achieved in such a short time period, I think, not only speaks to the power, like you said, of using technology and using live-streaming, but also, there is a massive demand. >> Gerritsen: There is a great need, yeah. >> For not only supporting, like from the perspective of the conference, you want to support and inspire and educate data scientists worldwide and support females in the field, but it really, I think, underscores, there is still in 2018, a massive need to start raising more profiles and not just inspiring undergrad females, but also reinvigorating those of us that have been in the STEM field and technology for a while. >> Gerritsen: That's right. >> So, what are some of the things, so, this year, not only are you reaching, hopefully about 100,000 people, you mentioned some of the countries involved today, but you also have a new first this year with the WiDS Datathon. >> That's right. >> Tell us about the WiDS Datathon, what was the idea behind it? You announced some winners today? >> Yeah. Yeah, so with WiDS last year, we really felt that we hit a nerve. Now there is an incredible need for women to see other women perform so well in this field. And, you know, that's why we do it, to inspire. But it's a one-time event, it's once a year. And we started to think about, what are some of the ways that we can make this movement, because it's really become a movement, into something more than just an annual, once-a-year conference? And so, Datathon is a fantastic way to do that. You can engage people for several months before the conference, and you can announce the winner at the conference. It is something that can be done really easily worldwide if it is supported again by the ambassadors, so the local WiDS organizations. So we thought we'd just try. But again, it's one of those things we say, "Oh, let's do it." We, I think, thought about this about six months ago. Finding a good data set is always a challenge but we found a wonderful data set, and we had a great response with 1100, almost 1200 people in the world participating. >> That's incredible. >> Several hundred teams. Yeah, and what we said at the time was, well, let's have the teams be 50% female at least, so that was the requirement, we have a lot of mixed teams. And ultimately, of course, that's what we want. We want 50-50, men-women, have them both at the table, to participate in data science activities, to do data science research, and answer a lot of these data questions that are now driving so many decisions. Now we want everybody around the table. So with this Datathon, it was just a very small event in the sense, and I'm sure next year it will be bigger, but it was a great success now. >> Well, congratulations on that. One of the things I saw you on a Youtube video talking about over the weekend when I was doing some prep was that you wanted this Datathon to be fun, creative, and I think those are two incredibly important ways to describe careers, not just in STEM but in data science, that yes, this can be fun. >> Yep. >> Should be if you're spending so much time every day, right, doing something for a living. But I love the creativity descriptor. Tell us a little bit about the room for interpretation and creativity to start removing some of the bias that is clearly there in data interpretation? >> Oh. (laughs) You're hitting the biggest sore point in data science. And you could even turn it around, you say, because of creativity, we have a problem too. Because you can be very creative in how you interpret the data, and unfortunately, for most of us, whenever we look at news, whenever we look at data or other information given to us, we never see this through an objective lens. We always see this through our own filters. And that, of course, when you're doing data analysis is risky, and it's tricky. 'cause you're often not even aware that you're doing it. So that's one thing, you have this bias coming in just as a data scientist and engineer. Even though we always say we do objective work and we're building neutral software programs, we're not. We're not. Everything that we do in machine learning, data mining, we're looking for patterns that we think may be in the data because we have to program this data. And then even looking at some of the results, the way we visualize them, present them, can really introduce bias as well. And then we don't control the perception of people of this data. So we can present it the way we think is fair, but other people can interpret or use little bits of that data in other ways. So it's an incredibly difficult problem and the more we use data to address and answer critical challenges, the more data is influencing decisions made by politicians, made in industry, made by government, the more important it is that we are at least aware. One of the really interesting things this conference, is that many of the speakers are talking to that. We just had Latanya Sweeney give an outstanding keynote really about this, raising this awareness. We had Daniela Witten saying this, and various other speakers. And in the first year that we had this conference, you would not have heard this. >> Martin: Really? Only two years ago? >> Yeah. So even two years ago, some people were bringing it up, but now it is right at the forefront of almost everybody's thinking. Data ethics, the issue of reproducibility, confirmations bias, now at least people now are aware. And I'm always a great optimist, thinking if people are aware, and they see the need to really work on this, something will happen. But it is incredibly important for the new data scientists that come into the field to really have this awareness, and to have the skill sets to actually work with that. So as a data scientist, one of the reasons why I think it's so fun, you're not just a mathematician or statistician or computer scientist, you are somebody who needs to look at things taking into account ethics, and fairness. You need to understand human behavior. You need to understand the social sciences. And we're seeing that awareness now grow. The new generation of data scientists is picking that up now much more. Educational programs like ours too have embedded these sort of aspects into the education and I think there is a lot of hope for the future. But we're just starting. >> Right. But you hit the nail on the head. You've got to start with that awareness. And it sounds like, another thing that you just described is we often hear, the top skills that a data scientist needs to have is statistical analysis, data mining. But there's also now some of these other skills you just mentioned, maybe more on the softer side, that seem to be, from what we hear on theCUBE, as important, >> Gerritsen: That's right. >> As really that technical training. To be more well-rounded and to also, as you mentioned earlier, to have to the chance to influence every single sector, every single industry, in our world today. >> And it's a pity that they're called softer skills. (laughs) >> It is. >> Because they're very very hard skills to really master. >> A lot of them are probably you're born with it, right? It's innate, certain things that you can't necessarily teach? >> Well, I don't believe that you cannot do this without innate ability. Of course if you have this innate ability it helps a little, but there's a growth mindset of course, in this, and everybody can be taught. And that's what we try to do. Now, it may take a little bit of time, but you have to confront this and you have to give the people the skills and really integrate this in your education, integrate this at companies. Company culture plays a big role. >> Absolutely. >> This is one of the reasons why we want way more diversity in these companies, right. It's not just to have people in decision-making teams that are more diverse, but the whole culture of the company needs to change so that these sort of skills, communication, empathy, big one, communication skills, presentation skills, visualization skills, negotiation skills, that they really are developed everywhere, in the companies, at the universities. >> Absolutely. We speak with some companies, and some today, even, on theCUBE, where they really talk about how they're shifting, and SAP is one of them, their corporate culture to say we've got a goal by 2020 to have 30% of our workforce be female. You've got some great partners, you mentioned Walmart Labs, how challenging was it to go to some of these companies here in Silicon Valley and beyond and say, hey we have this idea for a conference, we want to do this in six months so strap on your seatbelts, what were those conversations like to get some of those partners onboard? >> We wouldn't have been able to do it in six months if the response had not been fantastic right from the get-go. I think we started the conference just at the right time. There was a lot of talk about diversity. Several of the companies were starting really big diversity initiatives. Intel is one of them, SAP is another one of them. We were connected with these companies. Walmart Labs, for example, one of the founders of the company was from Walmart Labs. And so when we said, look, we want to put this together, they said great. This is a fantastic venue for us also. You see this with some of these companies, they don't just come and give us money for this conference. They build their own WiDS events around the world. Like SAP built 30 WiDS events around the world. So they're very active everywhere. They see the need, of course, too. They do this because they really believe that a changed culture is for the best of everybody. But they also believe that because they need the women. There is a great shortage of really excellent data scientists right now, so why not look at 50% of your population? >> Martin: Exactly. >> You know, there's fantastic talent in that pool and they want to track that also. So I think that within the companies, there is more awareness, there is an economic need to do so, a real need, if they want to grow, they need those people. There is an awareness that for their future, the long term benefit of the company, they need this diversity in opinions, they need the diversity in the questions that are being asked, and the way that the companies look at the data. And so, I think we're at a golden age for that now. Now am I a little bit frustrated that it's 2018 and we're doing this? Yes. When I was a student 30 some years ago, I was one of the very few women, and I thought, by the time I'm old, and now I'm old, you know, as far as my 18-year-old self, right, I mean in your 50s, you're old. I thought everything would be better. And we certainly would be at critical mass, which is 30% or higher, and it's actually gone down since the 80s, in computer science and in data science and statistics, so it is really very frustrating in that sense that we're really starting again from quite a low level. >> Right. Right. >> But I see much more enthusiasm and now the difference is the economical need. So this is going to be driven by business sense as well as any other sense. >> Well I think you definitely, with WiDS, you are beyond onto something with what you've achieved in such a short time period. So I can only imagine, WiDS 2018 reaching up to 100,000 people over these events, what do you do next year? Where do you go from here? (laughs) >> Well, it's becoming a little bit of a challenge actually to organize and help and support all of these international events, so we're going to be thinking about how to organize ourselves, maybe on every continent. >> Getting to Antarctica in 2019? >> Yeah, but have a little bit more of a local or regional organization, so that's one thing. The main thing that we'd like to do is have even more events during the year. There are some specific needs that we cannot address right now. One need, for example, is for high school students. We have two high school students here today, which is wonderful, and quite a few of them are looking at the live-stream of the conference. But if you want to really reach out to high school students and tell them about this and the sort of skill sets that they should be thinking about developing when they are at university, you have to really do a special event. The same with undergraduate students, graduate students. So there are some markets there, some subgroups of people that we would really like to tailor to. The other thing is a lot of people are very very eager to self-educate, and so what we are going to be putting together, at least that's the plan now, we'll see, if we can make this, is educational tools, and really have a repository of educational tools that people can use to educate themselves and to learn more. We're going to start a podcast series of women, which will be very, very interesting. We'll start this next month, and so every week or every two weeks we'll have a new podcast out there. And then we'll keep the momentum going. But really the idea is to not provide just this one day of inspiration, but to provide throughout the year, >> Sustained inspiration. >> Sustained inspiration and resources. >> Wow, well, congratulations, Margot, to you and your co-founders. This is a movement, and we are very excited for the opportunity to have you on theCUBE as well as some of the speakers and the attendeees from the event today. And we look forward to seeing all the great things that I think are going to come for sure, the rest of this year and beyond. So thank you for giving us some of your time. >> Thank you so much, we're a big fan of theCUBE. >> Oh, we're lucky, thank you, thank you. We want to thank you for watching theCUBE. I'm Lisa Martin, we are live at the third annual Women in Data Science Conference coming to you from Stanford University, #WiDS2018, join the conversation. I'll be back with my next guest after a short break. (upbeat music)