 Live from San Francisco, it's theCUBE. Covering Google Cloud Next 2018. Brought to you by Google Cloud and its ecosystem partners. Hey, welcome back everyone. We're live in San Francisco, CUBE coverage for Google Cloud Next 18. I'm John Furrier, Jeff Frick. Our next guest is Carol Carpenter, Vice President of Product Marketing here at Google Cloud and Naveed Arifani Adhimi. Welcome to theCUBE. Thank you very much. Thanks for coming on. Big data for good has been a topic we were just talking about here. Day three, what do you guys do and what's your relationship with Google? Because big data for good is really with cloud computing more relevant than ever before. Take a minute to explain your project. Sure, so in South Africa, we're a social nonprofit organization. We try and connect young people that have not employed, never employed to opportunities. And we are hosted in Google Cloud and we use GCP as our sole provider. And what we try and do is we use data to be able to understand young people, understand the facets and make a young person employable and match them to opportunities that we find. And so we describe opportunities and using different data points. And so all those data points that we have, we store them in Cloud SQL and we store them in BigQuery. And then we run analytics and matching to be able to find how these young people can contribute to the economy. How's it going so far? So far it's been great. It's allowed us to think about a 10x strategy. So when we're on-prem business, we were very limited by what we could provide bricks and mortar. And now we're looking and saying, well, how do we provide as much capacity and capability to these young people using self-service channels? So it really has just opened up the world of possibilities. And we're really looking at it and we're very excited because we've taken on some initiatives in Rwanda as well. And so we're taking on a global and Africa-wide kind of strategy, which I think without Cloud, we really wouldn't be able to do. So I wonder if you could just drill down because what are some of the data points that you look at and you measure? And is it identifying the data points and then finding the match? Or is it finding the critical ones that you really need to address as a priority to get kids to that position where they can get a job? I mean, it's really interesting because what we talk about, we talk about proxies for competence. So if you think about when you go and apply for a job, you kind of go and say, hello, here I am and I've done this job for so many years and that's your proxy for competence. So if you're a young person that just has a high school education and you're stepping in, we need to be able to describe you as a human, right? And so for those things, we look and say, what are your biographic information? What's your socialization? What kind of grit and energy to bring to the job? So we try and measure those things and we have as many contact points as we can get to be able to understand who is this individual really and use those data points and we have about 155 aspects that we use right now and that match them to different entry-level jobs. Yeah. So you're the enterprise architect of Herabee Youth Unemployment Accelerator. I love that term accelerator. That's right. And I also love the term enterprise architect because both are in need of some cloud, right? So one of the themes is digital transformation which is kind of a generic term. The analysts all talk about it but really we're talking about cloud mobile digital world and the power that can bring. Accelerator on the youth side, they need an app. So you're essentially providing a digital capability not the old brick and mortar. That's right. How do you architect all this? Because you got to assume there's an app at the edge either downloadable app or website, phone. So we actually have quite an interesting problem to solve because for our young people they don't have access to apps. The majority of our young people are on feature phones, basic phones, not smartphones and data in South Africa is very expensive. So for that young person we need to provide as low a touch a connection point to our services without making that cost them something, right? So we built a very basic Mobi site, no JavaScript as blank as you can get it's very boring if you look at it. The lightweight of Mobi. Very lightweight. But it's the tip of an iceberg and so from there we collect certain information but then we have an award-winning contact center that makes 35,000 calls every month and we engage with that young person on an outbound call for about 15 minutes and it's that 15 minutes that we use to talk to this young person, understand about them, figure out who they are and what they are and use that to gather our data points. We then have assessments that we run, so we run psychometric assessments, we're competence assessments and we gather all those data points and we start understanding this young person in a way that we can go to an employer because on that side for the employer we need to be able to say you trust us that when we give you this young person that we say this person will do well in your job well you have to have trust in us to be able to do that and so we need to provide that data to say well this is how we came up with it. So we take quite a lot of effort. You're verifying in a way you're putting your reputation on the line with the candidates. Yes. At the same time you don't know when the inbound touch is going to happen, so you're going to have all that material ready to go. That's right. That's where the big data kicks in. That's right. And so the big data, the collection of that information and the understanding of it and we're on a journey to start figuring out how can we use artificial intelligence, how can we use ML in a way that improves our accuracy but at the same time leaves out anything that may be biased toward these young people. So we're taking a very cautious approach to it but it's a lot of big data. We're trying to consume it as best we can plus we're trying to think about how do we provide provision our services for the employers because again it's a demand led business so we want to find as many jobs as we can so we can take young people to those jobs. So extend our reach to the employers and heavy lifting so that they don't have to. Yeah, so they don't have to, oh yeah. Carol, talk about the dynamic with Google Cloud because this is the theme we're hearing all week. You guys do the heavy lifting and at the edge of the user experience, you take the toil out of it. The word toil has been- It keeps coming up. It keeps coming up. Taking the toil, the hard work, friction out of it. In this case, the connectivity costs make being productive at that point of transaction. Exactly. They're doing the back end heavy lifting. This is kind of like a core theme across. That is what our promise, the promise of the cloud is supposed to be, right? Which is to remove all that back end toil. I love that word too. The toil, the mundaneness of it all so that folks like Haramby can actually focus on delivering great service to both potential employers and employees. So we're trying to automate as much of that infrastructure. That's what we announced a lot around, serverless around containers. This idea of you don't need to worry about it. You don't have to provision the server now. You don't have to worry about patches. You don't have to worry about security. We'll take care of that for you. I just love your phrase proxy for competence. Cause I can't help but think I've got kids college that you know, there's this whole, that's the whole objective of the application, right? And we've got SATs and PCTs and they take a couple of data sets, but relatively a number of data sets that you described. And I would imagine the kind of the intimacy of those data sets versus a couple of, you know, an ACT and SAT and a transcript. You probably have a really interesting insight and if you can correlate to the proxies of competency, now this is something that has a much greater kind of opportunity than just helping these kids that you need to help in is really important. But that's a really interesting take to use a much bigger data set, sophistication, great tools and infrastructure to do that mapping of competency to that job. Absolutely, and we're very focused on understanding how do we use this data to provision a network for our young people to be able to describe themselves and enter. So one of the things that we found in South Africa and I'm sure it's a fairly universal problem is that if you're unemployed, one of the things that prevents you from finding employment is you cannot access a network in which people that have jobs or describe jobs, you don't have access to their network. And so the ability to stand up and say, hey, this is who I am, these people have said, this is my profile as an individual. And say Harambee or whoever it is says that, I am competent in these things, that gives them an in, that gives them some way of entering that network. And for instance, we've done a certain study that said that if a young lady takes just a basic CV that has a stamp on it from Harambee with a description of who they are and what their competencies are, that improves their chances of finding a job by 30%, up to 30%, and that's significant, right? And this is not us finding the job for them, this is them going out and looking for a job. So it's describing and helping this person enter that network by providing, again, a proxy for competence. So about the relationship with Google, what is Google working with you guys on and what's next for you guys? So Google has helped us immensely. We've received those credits and those credits allowed us to take that first step into the cloud. It gave us a little bit of breathing room, all right? So we could take that step. We've also had access to some Googlers that have helped talk to us a little bit about ML and have been helping us out on that. And in terms of the next steps, it's 10x time, it's time to grow, it's time to use the scale, it's time to use the opportunity that we have to make the real impact that we've been searching for. And connect those jobs to those folks. Absolutely, because this is not a small problem, we've got a big problem to solve and we're really excited to be able to do it. Well, I'm glad you're doing that, it's a great, great mission. Kara, I want to get your thoughts finally just to kind of end the segment and kind of end our time here at Google Cloud. Good opportunity to talk to someone who's been looking at the landscape of the products. What's been the vibe of the show from your standpoint? Obviously you've been planning this for months. It's showtime, it's coming to a close. We're at day three, you heard, it's going to close in 30 minutes. Are you happy? You want? Yeah, I mean, we're thrilled, we're thrilled. We were just talking earlier, it's been a tremendous three days of just great interaction with fantastic customers, partners, developers, it's just the level of engagement. Google Cloud is about making the cloud available for everyone, we wanted this to be a place for people to engage, to make things, to try things, to be hands on, to be in sessions with people like Harambe, to actually understand what cloud can do. And we're super excited, we've seen that in spades, the feedback has been tremendous, I hope you've heard it as well. We're really excited. We believe that the capabilities we have around what we're doing in data analytics, machine learning, on top of this incredibly robust infrastructure, we really believe there are amazing problems we can solve together. We had a couple of our reporters here earlier saying people who think Google's far behind is not here at the event. I got to say, I give you guys some props, you guys are bringing, we know you got great technology, everyone kind of knows that, who knows Google, certainly knows the size and scope of the great technology, but you're making it consumable and you're thinking about the enterprise versus we're Google, use our great stuff, because we use it, you're like Google. People aren't like Google, because no one has that many servers. So it's self-awareness, this has really been a great stride, I think you guys have shown, and the customers on stage. I mean, that's the proof in the pudding. The end of the day. They're fantastic, showing how you can actually apply it, how you can apply AI, machine learning to actually solve real world problems, that's what we're most excited about. I mean, like you said, lots of great technology, what we want to do is connect the dots. And Diane Green, I thought of my favorite sound bite was, Securities number one, worry, AI is the number one opportunity. I think that if you look at it from that lens, kind of everything falls into place. Well, thanks for coming on, thanks for having theCUBE this week, Google. And congratulations on your great venture, and good luck with your initiative. Thank you very much. All right, it's theCUBE coverage here, live in San Francisco. I'm Trevor Jeff Frick, Dave Vellante. I went home last night, he's back in our office taking care of some business. I want to thank everyone for watching, and that's a wrap here from San Francisco. Thanks for watching.