 Live from Washington D.C., it's theCUBE. Covering Boomi World 19, brought to you by Boomi. Welcome back to theCUBE, the leader in live tech coverage. I'm Lisa Martin, John Furrier is my co-host and we are at Boomi World 2019 in Washington D.C. Very pleased to be joined by the founder of Boomi and the co-founder and CEO of Guru, Rick Nucci. Hey Rick. Hello. Welcome to theCUBE. Thanks for having me. This is very cool setup. Yeah, isn't it? Yeah. So this is founder of Boomi. It's pretty cool to have a celebrity on our stage. A celebrity. A celebrity. Tell, talk to us about all that back in the day, back in Philadelphia when you had this idea for what now has become a company that has 9,000 plus customers in 80 plus countries. Yeah, I am beyond proud of this team and just how well they have done and made this business into what it is today. Yeah, back, way back in 2007, we were really looking at the integration market and back then cloud was really like an unknown future. Right? It was creeping up the hype cycle of the Gartner. You know, hype cycle is like my favorite thing they do. A lot of people were dismissing it as a fad and we were early adopters of cloud internally at Boomi. We were early users of Salesforce and NetSuite and just thought, I made a bet and a lot of this stuff is luck as any founder will tell you, any honest founder will tell you and recognize that, hey, you know, if the world were to move to cloud, how would you actually think about the integration problem? Because it would be very different than how you would think about it in the on-prem days, right? When you have everything in your own data center and behind your own four walls, in this world, everything's different. Security's a huge deal. The way data moves and has to mediate between firewalls is a big deal and none of these products are built like this. And so really wanted as a team, and I remember these early conversations and had the willingness, I think, to take a big bet and swing for the fences. And what I mean by that is really build a product from the ground up in this new paradigm of cloud and kind of take a bet and say, look, hey, if cloud does kind of take off, this will be awesome for Boomi. If not, well, we'll be in the line of all the other startups that have come and gone. And I think we ended up in a good spot. Yeah, that's a great point, Rick, about the founders being honest. And a lot of it is hard work, but having a vision and making the multiple bets and big bets. I remember when EC2 came out, it was a startup dream too, by the way. You could just, you know, the person in the data center. But it wasn't fully completed. It was actually growing very fast. More services were coming on. They were web services. So that was API based concepts back then. When was the crossover point for you guys going, okay, we got this. The bets are coming in. We're going to double down. We're going to double down on this. What were some of those moments where you started to get visibility that was a good bet, and what did you do? Yeah, yeah, what it really was, was the rise of SaaS, very specifically, and the rise of business applications that were being re-architected in the cloud. And everybody knew about Salesforce. But there weren't a lot of other things back then, right? And there was NetSuite and a handful of others. But then you started to see additional business units start to build cloud. And you had in the HR space with success factors in Teleo and marketing automation space with Eloqua and Marketo, CRM space. We all know that story. E-commerce space, procurement, right? And you start to see these best of breed products rise up, which is amazing, but as that was happening, it was proliferating the integration problem, right? And so what became really clear to us, I think, as we were kind of going through this and finding product market fit for Boomi, again, back in 2007, 2008, that was the pattern that emerged. Like, hey, every time someone buys one of these products, they are going to have to integrate, because you're talking about employee data, customer data, you have to integrate this with your other systems. And that was going to create an opportunity for us. And that was where we were like, okay, I think we're on to something. You know, we've been doing theCUBE for 10 years. Okay, we made a big bet that people, authentic conversation would be a good bet. Turns out it worked. We love it. Things going great. But now we're living in a world now that's getting more complex. And I want to get your thoughts to Dave Vellante and myself, Stu, where we've been talking about how clouds changed. And we kind of were goofing on the Web 2.0 metaphor by saying, oh, cloud 1.0, cloud 2.0. But I want to get your thoughts on how you might see this because if you say cloud 1.0 was Amazon, compute, storage, at scale, cloud native, all started, pretty much started there. Pretty straightforward if you're going to be born in the cloud. And you could work with some things there. But to bring multi-cloud and for enterprises to adopt with this integration challenge, cloud 2.0 unveils some new things. Like for instance, network management now is observability. Configuration management is now automation. So you start to see things emerge differently in this cloud 2.0 kind of operating model. How do you see cloud 2.0? I mean, do you believe that, one, there's a cloud 2.0 the way I said it or, and if so, what is your version of what cloud 2.0 would look like? Yeah, I think, yes, definitely think things are changing. And the way that I think about it is that we're continuing to unbundle. And what I mean by unbundle is we're continuing to proliferate that buyers are willing to buy and therefore we're continuing to proliferate relatively narrower and narrower and deeper and deeper capabilities and functionalities. And one big driver of that is AI, specifically machine learning. And not the hypey stuff, but the real stuff. It's funny, man, when you compare right now AI and what I was just talking about, it's the same thing all over again, right? It's hype cycle crawling up the thing. Okay, so, but now I think the recipe for good AI products that really do solve problems is that they're very intentionally narrow and they're very deep because they're gathering good training data and they're built to solve a very specific problem. So I think- Like domain expertise, domain specific. Exactly. Industry expertise, domain expertise, use case. If you're gathering training data about a knowledge worker, the data you'll gather is very different if you're a salesperson or an HR professional or an engineer, right? And I think the AI companies that are getting it right are really dialed in and focused on that. So as a result, you see this proliferation of things that might be layered on top of big platforms like CRMs but, and technologies like Slack, which is creating a place for all this to come together, but you're seeing this unbundling where you're getting more and more kind of like, almost microservices, not quite, but very fine-tuned specific things kind of coming together. So machine learning, I totally agree with you. It's definitely hype, but the hardcore machine learning has kind of a math side to it and a cognition side, cognitive learning thing. But also data is a common thread here. I mentioned domain specific. All about the data. So if data is super important, you want domain expertise, which I agree with, but also there's now a horizontal scalability with observation data. The more data you have, the better and machine learning may or may not, depending on what the context is. So you have contextual data. This is a hard thing. What's your view on this? Because this is where people kind of maybe get caught around the axis of machine learning hype and not really narrow in on what their data thinking is. 100%. What's your score? 100%. I think people will tend to fall in the trap of focusing on the algorithms that they're building and not recognizing that without the data, the algorithms are useless, right? And that it's really about how, as a ML problem that you're trying to tackle, are you gathering data that's good, high quality, scalable, accurate, protected and safe? Because now, for different reasons, but again, just like when we were moving to cloud, security and privacy are up most important because for any AI to do its job well, it has to gather a lot of data out of the enterprise and store it and train off of it. It's interesting, a lot of the cloud plays, we see us was just a unicorn right out of the gate and they were a pioneer is what it is. They were cloud before cloud was cloud, as we know it today. But you see a lot of things like the marketing, automation, cloud platform, it's a marketing cloud, I got a sales cloud, almost seem too monolithic. And you see people trying to unbundle that I think you're right or break it apart because the data is like stuck in this full stack model because if you agree with what you said, it's horizontal scalability and vertical integration is the architecture. Technically that's half stack. Half stack developers are valuable now. Totally and yes, I like that term. The other problem that I think you're getting at is like tendency isolation of that data. A lot of things were built with that in mind meaning that the best AI you're going to build is only going to be what you can derive from one customer set of data, right? Whereas now people are designing things intentionally such that the more people, the more customers that are using the thing, the better and smarter it gets. And so to your point about monolithic, I think the opportunity that the startup, the next wave of startups have is that they can design in that world and that just means that their technology will get better faster because it'll be able to learn from more data. You know, the sentiment changes a lot in cloud. I want to get your thoughts because you guys at Boomi here are on a single tenant instance model because the collective intelligence of the data benefits everybody as more people come in. That's a beautiful flywheel. It feels a lot like Amazon model to me. But the old days, multi-tenancy was the holy grail. Maybe that came from the telcos or whatever hosting world. What's your view on single tenant instance on a SaaS business versus say multi-ten, this trade-offs and pros and cons, what's your opinion? And what do you lean on this one? Yeah, I mean, we both Boomi and Guru, so two eras worth or whatever, you have to have some level of tendency isolation for some level of what you do, right? And at Boomi what we did is we separated the sensitive private data. You know, Boomi has customers processing payroll through its product, right? So very, very sensitive stuff. Absolutely has to be protected and isolated per tenant and Boomi and Guru is signing up for that in the clauses that we signed to with security agreements. But what you can decouple from that is more of the metadata or the attributes about that data and that customer, right? So Boomi you were referring to launched way back when Boomi suggests which basically learned, as all the people were building data maps connecting different things together, Boomi could learn from all that and go, oh, you're trying to do this? Well, these, however many other customers, let me suggest how these maps are drawn, right? And Guru, we're following a very similar pattern. So Guru, we stored knowledge which is also tends to be IP for a company. And so yes, we absolutely adhere to the fact that only a handful of our employees can ever see that stuff and that's because they're in DevOps and they need it to keep things running, right? But all the tenants are protected from one another. No one could ever leak to another one. But there are things about organization and structure and tagging and learnings you can get that are not that sensitive stuff that does make the product better from a AI perspective, the more people that use it. And so I don't know that I'm giving you like I'll one or another, but I think it does come down to how you intentionally design your data. The decoupling is the critical piece. Absolutely. This is the cloud architecture. Yes. Decouple, use APIs to connect. Highly cohesive elements and the platform can be cohesive if shared. Absolutely. And you can still get all the benefits of scalability and elastic growth and yeah, 100%. Along that uncoupling line, tell us a little bit briefly about what Guru is and then I want to talk about some of the use cases. I know I'm a big Slack user. You probably are too, John. How, talk to us about what you're doing there but just give our folks a sense of what Guru is and all that good stuff. Sure, I mean Guru's, you know, in some ways like Boomi, rethinking a very old problem. In this case, it's knowledge management. That's a concept we've talked about for a long time and I think these days, it has really become something that is, does impact a company's ability to scale and grow reliably. So very specifically what we do is we bring the knowledge that employees need to do their job to them when they need it. So imagine if you're a customer support agent and you're supporting Spotify. You're an employee of Spotify and I write in and I want to know about the new Hulu partnership. As an agent, you use Guru to look up and give me that answer. And you don't have to go to a portal. You don't have to go to some other place to do that. Guru's sitting there right next to your ticket or your chat as you're having it in real time saying, hey, they're asking about Hulu. These are the important things you want to know and talk about. And then the other half of that is we make sure that that doesn't go stale. The classic problem with knowledge products is the information, when you're talking about something like product knowledge changes all the time. And the world we live in is moving faster and faster and faster. So we used to ship product once a year, once every two years. Now we ship product every month, sometimes a couple of times a month. Can we get a Guru bot for our journalism and our Cube hosts? We can be real time. I would be great. I would be great. I'm a Guru journalist. Actually, yeah, but does that right in your ear there? I'll take it. Just prompting you with exactly. So and then you asked about Slack. That's a really great partner for us. They were an early investor in the company. They're a customer. But together, if you think about where a lot of knowledge exchange happens in Slack, it's, hey, I need to know something. I think I can go Slack, John, because I think he'll know the answer. He knows about this. And you're like the 87th person who's asked me that same thing over again. Well, with Guru being integrated into Slack, you can just say Guru, give them the answer and you don't have to repeat yourself. And that expert fatigue problem is a real thing. That's a huge issue. And as your company grows and more and more people are, oh, poor John's getting buried for being the expert, right? That one of the reasons he got you there now, he's getting burned out and buried from it. And so we seek to solve that problem. And then post-Guru, a company will scale faster. They'll onboard their employees faster. They'll launch products better because everyone will know what to talk about when. They'll frequently ask questions operating system. Exactly. That's a great analogy, right? And making it living, right? Because all those efficacy change all the time. That's the important part too, is keeping it relevant. Yes. 24 by seven. Absolutely, yeah. Contextual data analysis is really hard. What's the secret sauce? The secret sauce is that we live where you work. The secret sauce is that we focus very specifically on specific workflows like that customer support agent. And so by knowing what you're doing and what ticket you're working on and what chat you're having with a customer, Guru can be anticipatory over time and start to say, hey, you probably want to talk to him about this and bring that answer to you. It's because we live where you work. And that was frankly accidental in a lot of ways. We were trying to solve the problem of knowledge living where you work. And then what we realized is, wow, there's a lot of interesting stuff that we can learn and give back to the customer about what problems they're solving and when they're using Guru and why. And that only makes the product better. So that's really, I think the thing that if you ask our typical customers really gets them kind of like excited. They'll say like, hey, because of Guru, I feel more confident when I'm on the phone that I'm always going to give the right answer. That's awesome. I love hearing customers talk about or even business leaders talk about some of the, I want to say accidental discoveries or capabilities. But just how over time, more and more and more value gets unlocked if you could actually really extract value from that data. Last question, Rick, I need to know what's in a name? The name Bumi, the name Guru. Yes. Well, I'll start with the less exciting answer, which I always get asked about, which is Bumi, which is a Hindi word that means earth or from the earth. And sometimes if you're ordering it, say at Indian restaurant, you'll see BHOMI and that might be the vegetables on the menu, right? That name came from an early employee of the company. I wish I could say that it had a connection to the business. It really doesn't, it just was like, it looks cool and people tend to remember the name. And honestly, like there have been so many moments in the early, early days where we're like, should we change the name? It doesn't really, and we're like, no, what? People tend to, it sticks with them. It's kind of exciting, you know, and we kept it. Guru on the flip side, one of our early employees came up with that name too. And I think she was listening to me talk about what we were doing and she's like, oh, like that thing is like a Guru to you. And so the brand promise is that you feel like a Guru in your area of expertise within a company and that our product plays a relatively small role in you having that expert, you know, feeling confident about that expertise. I love that, awesome. Rick, thank you so much for joining John and me on theCUBE today. We appreciate it. Thank you. Thanks. For John Furrier, I'm Lisa Martin. You're watching theCUBE from Boomi World 2019. Thanks for watching.