 From around the globe, it's theCUBE with digital coverage of UiPath Live, the release show. Brought to you by UiPath. Hi everybody, this is Dave Vellante. Welcome back to our RPA Drill Down. Ted Kumert is here, he's the executive vice president for products and engineering at UiPath. Ted, thanks for coming on, great to see you. Dave, it's great to be here, thanks so much. You have your background, it's pretty interesting. You started as a Silicon Valley engineer, they pulled you out and did a huge stint at Microsoft. You got experience in SaaS, you got VC chops with Madrona. And at Microsoft, I mean, you saw it all, the NT, the CE space, workflow, I mean, even MSN, you did stuff with MSN and then the all important data. So I'm interested in what attracted you to UiPath. Yeah, Dave, I feel super fortunate to have worked in the industry in the span of time. It's been an amazing journey. And I had a great run at Microsoft, it was fantastic. You mentioned one experience in the middle there. When I first went to the server business, the enterprise business, I owned our integration and workflow products. And I would say that's the first I encountered this idea. And often in the software industry, there are ideas that have been around for a long time. And what we're doing is refining how we're delivering them. And we had ideas we talked about in terms of business process management, business activity monitoring, workflow, the ways to efficiently able somebody to express a business process in a piece of software, bring systems together, make everybody productive, bring humans into it. These were the ideas we talked about. Now, in reality, there were some real gaps because what happened in the technology was pretty different from what the actual business process was. And so let's fast forward then. I'm at Madrona Venture Group, Seattle based venture capital firm. And we actually made a decision to participate in one of UI pass fundraising rounds. And that's the first I really came encounter with the company and had to have more than an intellectual understanding of RPA. Because when I first saw it, I said, oh, I think that's desktop automation. I didn't look very close. Maybe that's going to run out of runway, whatever. And then I got more acquainted with it. And I figured out, oh, there's a much bigger idea here. And the power is that by really considering the process and the implementation from the humans work in, then you have an opportunity really to automate the real work. Not that what we were doing before wasn't significant. This is just that much more powerful. And that's when I got really excited. And then the companies, you know, the company's statistics and growth and everything else that speaks for itself in terms of an opportunity to work. I believe in one of the most significant platforms going in the enterprise today and work at one of the fastest growing companies around. It was like almost an automatic decision that's decided to come to the company. Well, you know, you bring up a good point is you think about software, you know, historically through our industry, a lot of it was, okay, here's the software. Now figure out how to map your processes to make it all work. And then today the processes, especially, you know, you think about this pandemic, the processes are unknown. And so the software really has to be adaptable. So I'm wondering, and essentially we're talking about a fundamental shift in the way we work. And is there really a fundamental shift going on and how we write software and how would you describe that shift? Well, there certainly are. And that's in a way that's the job of what we do when we build, you know, platforms for the enterprises is try and give our customers a new way to get work done. That's more efficient and helps them build more powerful applications. And that's exactly, you know, what our PA does like the efficiency, you know, it's not that this is the only way in software to express a lot of this, it just happens to be the quickest, you know, in most ways, especially as you start thinking about initiatives like our StudioX products to, you know, what we talk about is enabling citizen developers. You know, it's an expression that allows customers to just do what they could have done others much more quickly and efficiently. And the value on that is always high, certainly in an unknown era like this, it's even more valuable. I mean, there's specific processes we've been helping automate in the healthcare, in financial services with things like SBA loan processing that we weren't thinking about six months ago or they weren't thinking about six months ago. You know, we're all thinking about how we're reinventing the way we work as individuals and corporations because of what's going on with the coronavirus crisis and, you know, having a platform like this that gives you agility and then mapping the real work to what your computer state and applications all know how to do is even more valuable in a climate like that. Yeah, I mean, what attracted us originally to UiPath and new Bobby, Patrick, CMO and he said, Dave, go download, you know, a copy and go build some automations and go try it with some other companies. And so that really struck us as wow, this is actually quite simple. And yet at the same time, and so you've of course been automating all these simple tasks, but now you've got real aspiration. You know, you're sort of glomming on to this term of hyper automation. You've made some acquisitions. You've got a vision that really is taking you beyond kind of paving the cow path, I sometimes say, of all these existing processes. It's really trying to discover new processes and opportunities for automation, which you would think after 50 or whatever years we've been in this industry, we've attacked a lot of it, but wow, we have seems like we have a long way to go. Again, especially what we're learning through this pandemic. Your thoughts on that? Yeah, I'd say hyper automation. And it's actually a Gartner term. It's not our term. And but there is a bigger idea here built around the core automation platform. So let's talk for a second, just what's not about the core platform and then what really hyper automation really means around that. And I think of that as the bookends of, you know, how do I discover and plan? How do I improve my ability to do more automations and find the real opportunities I have? And then how do I measure and optimize? And that's a lot of what we delivered in 20.4 is new capability. So let's talk about discover and plan. One aspect of that is kind of the wisdom of the crowd. And we have a product we call Automation Hub that is all about that. Enabling people who have ideas, they're the ones doing the work, they have the observation into what efficiencies can be, enabling them to either with our past capture utility, capture that and document that or just directly document that. And then people across the company can then collaborate eventually on building the best ideas out of that. So there's capturing the crowd and then there's a more scientific way of capturing actually what the opportunities are. And so we've got two products we introduced. One is process mining. And process mining is about going outside in from the, let's call it the larger processes, more end to end processes in the enterprise. Things like order to cash and procure to pay and helping you understand by watching the events and doing the analytics around that, where are your bottlenecks, where are your opportunities? And then task mining said, let's watch an individual or group of individuals, what their tasks are, let's watch the log of events there, let's apply some machine learning processing to that to say, here's the repetitive things we found and really helping you then scientifically discover what your opportunities are. And these ideas have been around for a long time, process mining is not new, but the connection to an automation platform we think is a new and powerful idea and something we plan to invest a lot in going forward. So that's the first bookend. And then the second bookend is really about attaching rich analytics. So how do I measure it? So there's operationally, how are my robots doing? And then there's everything down to return on the desk. How do I understand how they're performing versus what I would have spent if I was continuing to do them the old way? Yeah, that's big. The hero reports for the executives to say, hey, this is actually working, but at the same time, you've got to take a systems view, right? You don't want to just optimize one part of the system at the detriment to others. And that's why, so you talk about process mining, which is kind of, you know, discovering the backend systems, ERP and the like where the task mining, it sounds like it's more the sort of collaboration and kind of front end. So that whole system thinking really applies, doesn't it? Yeah, it's very much so. And, you know, another part of what we talked about then in the system is, you know, how do we capture the ideas and how do we enable more people to build these automations? And that really gets down to, we talk about it in our company level vision as a robot for every person. Every person should have a digital assistant. It can help you with things you do less frequently. It can help you with things you do all the time to do your job. And how do we help you create those? And we've released a new tool we call StudioX. So for our RPA developers, we have Studio. And StudioX is really trying to enable a citizen developer. It's not unlike the art that we saw in business intelligence. There was the era where analytics and reporting were the domain of experts and they produced formalized reports that people could consume. But the people that had the questions would have to work with them and couldn't do the work themselves. And then along comes, you know, ClickView and Tableau and Power BI enabling the self-service model. And all of a sudden people could do that work themselves and that enabled powerful things. We think the same arc happens here and StudioX is really our way of enabling that, you know, that citizen developer with the ideas to get some, you know, automation work done on their own. Got a lot in this announcement. Things like document understanding. You got, you kind of bring your own AI with AI fabric. How are you able to launch so many products and have them, you know, fit together? You've made some acquisitions. Can you talk about the architecture that enables you to do that? Yeah, it's clearly in terms of ambition. And I've been there for 10 weeks, but in terms of ambition, you know, and have to have been there when they started the release after four or three in October to know that this is the most ambitious thing that a company has ever done from a release perspective. Just in terms of the surface area we're delivering across now as an organization is substantive. You know, we talk about, you know, thousand feature improvements, you know, hundreds of discrete features and, you know, new products as well as, you know, now our automation cloud has become generally available as well. So we've had muscle building, you know, over this past time to become world-class at offering SaaS in addition to on-premises. And then we've got this big surface area. And, you know, architecture is a key component of how you can do this. You know, how do you deliver efficiently, you know, the same software on-premises and in the cloud? Well, you do that by having the right architecture and making the right bets. And certainly you look forward how our company's doing this today. It's really all about cloud-native platform. But it's about an architecture, you know, such that we can do that efficiently. So there is a lot about just your technical strategy and then it's just about kind of discipline and customer focus. It keeps you focused on the right things. You know, StudioX was a great example of, you know, we were led by customers to a lot of what we actually delivered. A couple of the major features in it, you know, certainly the out-of-box templates, the studio governance features, where it came out of customer suggestions, I think we had about a hundred that we have sitting in the backlog, a lot of which we've already done, and really being disciplined and really focused on what customers are telling. So make sure you have the right technical strategy and architecture really follow your customers and really stay disciplined and focused on what matters most as you execute on the release. What can we learn from previous examples? I mean, I think about, for instance, SQL Server. You obviously have some knowledge in it. You know, kind of started out, you know, pretty simple workloads. And then at the time we all said, wow, it's a lot, it's a lot more powerful to come from below than it is if a DB2 or an Oracle to sort of go down market. And Microsoft proved that and obviously built in the robustness necessary. Is there a similar sort of metaphor here with regard to things like governance and security, just in terms of where UI path started and where you see it going? Well, I think the similarities have more to do with just, we have an idea of a bigger platform that we're now delivering against. And, you know, in the database market, that was we started, SQL Server started out as more of just a transactional database product and ultimately grew to all of the workloads in the data platform, including, you know, transaction for transactional apps, data warehousing, and as well as business intelligence. And I see the same analogy here of thinking more broadly about the needs and what the ability of an integrated platform, what it can do to enable great things for customers. I think that's a very, you know, a very consistent thing. And I think another consistent thing is know who you are. SQL Server knew exactly who it had to be when it entered the database market, that it was going to set a new benchmark on simplicity, TCO, and that was going to be a way it differentiated. In this case, we're out ahead of the market. You know, we have a vision that's broader than where a lot of the market is today. I think we see a lot of people coming into the space, but we see them building to where we were and we're out ahead. So we are operating from a leadership position. And I'm not going to tell you one's easier than the other. And both you have to execute with great urgency, but we are really executing out ahead. So we've got to keep thinking about, and there's no one's tail lights to follow. We have to be the ones really blazing the trail on what all of this means. I want to ask you about just the incorporation of sort of existing systems. Some markets, they take off. It's kind of a one-shot deal and the market just embeds. I think you guys have bigger aspirations than that. I mean, I look at it like a service now. Misunderstood early on, built the platform, and now really is a fundamental part of a lot of enterprises. Will, and I also look at things like EDW, which again, you have some experience in. In my view, it failed to live up to a lot of its promises, even though it delivered a lot of value. You look at some of the big data initiatives. You know, EDW still plugs in, it's the system of record. Okay, that's fine. How do you see RPA evolving? Are we going to incorporate, do we have to embrace kind of existing business process systems, or is this largely a do-over in your opinion? Well, I think it's certainly about a new way of building automation and it's starting to incorporate and include the other ways. And Matt, you know, for instance, in the current release, we added support for long-running workflow. This is about human workflow-based scenarios. Now the task is, you know, now the human is collaborating with the robot and, you know, we built those capabilities. And so I do see us combining some of the old and new way. I think one of the most significant things here is also that, you know, the impact that AI and ML-based technologies and skills can have on the power of the automations that we deliver. And, you know, we've certainly got a surface area that, you know, I think about our AI and ML strategy in two parts, that we are building, you know, first-class, you know, kind of first-party skills that we're including in the platform and then we're building a platform for third-parties and customers to bring there but their data science teams have delivered. So they, those can also be a part of our ecosystem and part of automations. And so things like document understanding, you know, how do I, you know, easily extract data from more structured, semi-structured and completely unstructured documents, you know, accurately, and include those in my automations. Computer vision, which gives us an ability to automate at a UI level across other types of systems than, say, a Windows in a browser-based, you know, application. And task mining is built on a very robust, multi-layer ML system. And the innovation opportunities, I think, just consider there, you know, continue there. You think it's a macro level if there's aspects of machine learning that are about captured human knowledge. Well, what exactly is an automation but captured in a way you're capturing a lot of human knowledge, the impact of ML and AI is going to be significant going out in the future. Yeah, I want to ask you about that. I mean, I think a lot of people just are afraid of AI, like it's a separate thing and they got to figure out how to operationalize it. And I think companies like UiPath are really in a position to embed UI into applications, AI into applications everywhere so that maybe those folks that haven't climbed on the digital bandwagon, you know, who are, I think now with this pandemic are realizing, wow, we better accelerate this. They can actually tap machine intelligence through your products and others as well. Your thoughts on that sort of narrative. Yeah, I agree with that point of view. It's, AI and ML is still a maturing discipline across the industry and you have to build new muscle and you build new muscle in data science. And it forces you to think about data and how you manage your data in a different way. And that's a journey we've been on as a company to not only build our first-party skills but also to build the platform. It's what's given us the knowledge that to help us figure out, well, what do we need to include here so our customers can bring their skills actually to our platform. And I do think this is a place where we're going to see the real impact of AI and ML in a broader way based on the kind of apps it is and the kind of skills we can bring to bear. Okay, last question. You're 10 weeks in when you're, I don't know, 50, 100, 200 weeks in. What should we be watching? What do you want to have accomplished? Well, we're listening, we're obviously listening closely to our customers. I mean, right now we're still, we're having a great week because there's nothing like shipping new software. So right now we're, we actually are thinking deeply about where we're headed next. We see there's lots of opportunities and robot for every person in that initiative. And so we've launched a bunch of important new capabilities there and we're going to keep working with the market to understand how we can, how we can add additional capability there. We've just got the GA of our automation cloud. I think you should expect more and more services in our automation cloud going forward. I think this area we talked about in terms of AI and ML and those technologies, I think you should expect more investment and innovation there from us and the community, helping our customers. And I think you will also see us then, as we talked about this convergence of the ways we bring together systems to integrate and build business process. I think we'll see a convergence into the platform of more of those methods. So I look ahead to the next releases and want to see us making some very significant releases that are advancing all those things and continuing our leadership and what we talk about now is the hyper automation platform. Well Ted, a lot of innovation opportunities and of course everybody's hopping on the automation bandwagon. Everybody's going to want to PC your RPA hide zone and you're in the lead. So we're really excited for you. We're excited to have you on theCUBE. So thanks very much for all your time and your insight, really appreciate it. Yeah, thanks Dave. Great to spend this time with you. All right, thank you for watching everybody. This is Dave Vellante for theCUBE in our RPA drill down series. Keep it right there, we're right back right after this short break.