 Welcome back to theCUBE's live coverage, day two of theCUBE's wall-to-wall coverage, our state conference, 2023. I'm John Furrier, my co-host, Dave Vellante. We have a very special guest, Jeremy Burton, CEO of Zervink. He's on the first ever CUBE, by the way, in 2010. One of our guests, Jeremy, you're going way back 13 years ago. Yeah. You still got it, see these great hands? You're getting old, man. CMO of EMC, all those great years, you did amazing work. Now you're in startup observability space on the border snowflake. A lot of action. You're riding this new wave with Zervink, where it's the cloud native, it's the data area, it's the whole new stuff, but yet operating, trying to operate on that. So congratulations on the startup. Give us the update, what's going on? Yeah, I mean, we're still hiring. That's a big deal these days. A lot of startups going out of business, a lot of hiring freezers, but now we're doing quite well. We took us four years to build the product. I think the days, certainly in enterprise tech, I don't know consumer, you can maybe get lucky in consumer, but in enterprise tech, the barrier to entry on these markets is quite high. And so we've been building the product for four and a half years now. And I'd say in the last maybe six or nine months, we've really started to get it to where we wanted it to be. Hired a great head of engineering out of Stripe. You know, G-Gen was the VP of payments engineering at Stripe, and he's been working on the finishing touches. So Stripe do this great developer experience. It's frictionless, it's easy. And I think even with enterprise software, you've got to get things to spread virally. So yeah, observability, the space itself is heating up big time because the data volumes are out of control. It's interesting. You mentioned enterprise software and how long it takes to build the product. Four years is a lot of time. I want to ask you about what's kind of changed for the better or not so good for you. And at least there's a trend line there. But also what we're finding at Open Source at KubeCon and CNCF and the Linux Foundation, the developers are the new consumers. And we're calling it the B to D to B, the business to developer. If the developers like it, it's consumed and it becomes adopted. And that has nothing to do with marketing. That has to do with marketing in a way that doesn't feel like marketing. So if the developers like a product, if it helps their job, it's easier and it works, there's clearly all the VCs are looking at that as a metric on the investment side and also on the traction side. So combining that ease of use and then doing all the heavy lifting on the code, this is the new reality. So can you unpack that a little bit on the software side and then expand more on this developer adoption dynamic? Yeah, it used to be the case. I mean, back when I started my career back at Oracle in the mid-90s, we had a great sales team. And at times, there's certain versions of Oracle, like Oracle 6, for example, like not that great, but great sales team, you could sell it to senior people and then sort of force it on the technical folks in the organization. That you can't do that anymore, right? I think certainly you've still got to sell to the people who've got money in the budgets, but you can't jam it down the throats of the software engineers because you're going to get tissue rejection. And so even enterprise software now, I'd say it's got to be much slicker than it ever was in the past. There's got to be less friction. It's got to be geared towards a great developer experience. If you do that, people will pick it up, start using it. That's how you get the, I'd say the real revenue growth because the revenue is growing through usage without you having expensive sales people going to everybody individually. And on the code, it takes to build four years. Obviously it's a hardened product, complex, observability. What's been a tailwind for you and what's been different that you had to make a change or adjustment on? Yeah, I think, I mean, obviously people build apps differently these days, right? Forever it felt like we had the sort of Java application against, you know, pick the day, you know, Oracle database or whatever and the sort of Java monolith seemed to be around forever, but now you've got distributed apps. Distributed apps in some respects are more complex, but I think what's been key is continuous delivery has meant that developers could push new functionality into production more quickly than ever before, so you've got more change in production. And that's been the biggest driver for observability because you think about it, you see unknown problems in production every day. You can't do six weeks or six months of acceptance testing, figure out all the failure conditions, set up all your monitors and set up a nice dashboard to look for them, you can't do that anymore. You see new failures every day and the analogy I always give is, you know, hey, imagine, you know, we put a blindfold on you and we dropped you in San Francisco. Got to be pretty careful where I drop you these days because it might not be a great experience, but I take the blindfold off, what would you do? Right, you'd be like, okay, where's Twin Peaks? You know, where's the Salesforce Tower? You'd look around for data points that would give you context for your current location and it's exactly the same with observability. You dropped into a new scenario, I have a problem, where am I? What service has the issue? So what's the state of play today because you have companies that were doing metrics or whatever, you know, sort of legacy companies that now say, hey, we're observability too, you got, you guys like yourself had it, you know, you said it took four years, you had a less mature stack and so you got sort of the modern play and then there's this big mishmash of folks. Yeah, we're observability too, so how should we think about that? Yeah, the big companies are good at latching under the hot words, you know, hot words. And I mean, I know that because I was that one, right? And you remember, I mean, the big data, cloud meets big data. The hard thing is though for a big organization is to solve the problem in an elegant way that they often have to start again. Big companies, they don't start again because they've got incumbent businesses that have revenue associated with them and there's earnings for Wall Street and so that's where the startup has the advantage and so when we set out to build Observe, we didn't think about just logs or just tracing or just metrics, we just, we thought about, well, how do we build a system that can ingest any kind of unstructured event, whether it's a log event, a trace event, a time series metric or events coming out of salesforce.com or out of Service Now or wherever those events may be, how can we just suck them into one big snowflake database? I mean, that's another change because snowflake didn't really exist when folks like Datadog and Splunk and all these folks were around, they couldn't build on top of snowflake because it didn't exist. So we took advantage of new disruptive technology, Snowflake, and we thought, why can't we just build one big event database, stream everything in there, then we can start to ask questions that go across these silos, you know? And that's really where the value in the product and observability is. I see an alert over here that tells me a metric has been exceeded. Okay, where are the logs for that user session? Okay, now show me the traces and you don't actually have to leave the system in order to see all of the relevant contexts that you need to figure out root costs. So it's kind of an adjacency to security. I'd like you to help us understand where it fits. Because you seem like data protection, right? You see some data protection companies saying, yeah, now we're a security company. It's like, maybe not, I don't think you're repositioning as a security company, but you're an adjacency, are you not? Yeah, it's getting hard to tell. I mean, there are certain companies in our space that set out to build an observability tool. So like LightStep, that service now acquired, or Honeycomb, they're probably never going to do like security, why? Because they built an observability tool. The way we built Observe, it was quite what I call unopinionated in that we can accept observability data from Kubernetes clusters or AWS or from whatever application, but we can also accept security log events. So things coming out of Palo Alto Networks firewalls or identity management systems. It's just another event. And so the smaller companies that we started selling to when we first had a product, it turns out the DevOps team is actually the SecOps team. It's usually one or two people, right? And they're like, oh yeah, security, we do that too. Hey, by the way, can we ingest this stuff and can we pull out IP addresses? And we say, well yeah, you can do that. They're like, okay, great. And I'd say about 20% of our customers are now doing this kind of thing so much so that we were like, okay, maybe we should productize this. So we actually have a threat intelligence product at Observe, it's brand new. I don't even think we've announced it yet, but there you go. There we go, it's exclusive, exclusive. But we hired the guy that did Splunk Enterprise Security. He might know a thing or two. And he's produced this threat intelligence module so folks can take sort of black listed IP addresses from external open source feeds and correlate what they see externally with IP addresses that are sitting in their log files internally. And I sort of joke that we're building a sim for people who don't want a sim, which seems like that's almost everybody here. It's okay, so it's a product, you're saying. It is, we call it an app, but yeah, think of it as a product. It's an app on the platform. Okay, let's get into the trend, the data. As you are on the right side of this trend, I would call that, and I would say that you guys have pretty much made good calls. Where is the market relative to the people who aren't making the right calls? You and I both talked a lot, I think two times ago you were on theCUBE, how overfunded the sector was with the observability. We're seeing Splunk, for instance, out there struggling right now. And everyone kind of knows they're talking for private equity, and that's a real sad thing that Splunk had a good run going on there. Just overpriced, but didn't pivot properly into the trend line. There's going to be a lot of that, Jeremy. So how do you see the size of the street, kind of the ones who can't make it, the ones that cross over to the right side of history of this, what do I call it, platform, distributed computing, whatever you want to call it, cloud native, new modern app, centric, open AI, kind of. Like, I mean, there's a whole another tsunami coming. Yeah, yeah, who makes it, who doesn't? No, look, I think in tech we love fashion, right? We love a new trend and a new bubble, and I sort of joke that crypto was the last bubble that burst, and all those crypto companies are dead, and in my opinion, I don't think they're coming back. But the GPT thing came along, and all of a sudden, we're like, hey, you know, AI. We have a few GPT. So AI, I sort of think it's like the second coming of AI, particularly in the space that we're in. The first generation of AI, I think, yeah, it was great for identifying pictures of cats on the internet, and that was all cool. In the observability space, I was not a fan of AI because I didn't think it worked. I think it was, I thought it was sort of snake oil because these apps, modern apps, at least, they're changing every day. So you can't build a model that's worth anything because you're going to get lots of outliers. And so it didn't really have a lot of good applicability for modern distributed apps that change quite often. I think GPT comes along, and you think, oh, wait a minute here. It's not at the stage where you can just throw terabytes of machine data at an LLM, and it'll tell you the root cause of a problem. But you can feed it lots of great technical information. You can, I mean, we feed it code examples from our language Opal, and it can write code for you. It can help you get up to speed on a new product very quickly. If you see an error condition, it might not tell you root cause, but it can tell you the series of steps that you need to take in order to figure out what the root cause is. So I think this could improve productivity of this kind of tooling by 20 or 30%. You've seen Moneyball, and you were talking about Moneyball all the time on the keyboard, big data was around, but Dave knows I love to quote Moneyball. But the one scene at the end where Billy Bean's meeting the owner of the Red Sox, and the owner of the Red Sox says, it's always the first guy through the wall that's bloody, you know, and referring to him as the Moneyball. And then he says anyone who's not rebuilding their teams based upon the model is going to be dinosaurs, right? And so you got a lot of that going on here where you can see obvious fashionable trend lines that are structural, using data fast, being more agile with cloud scale. It's pretty obvious, and as business try to figure that, what's your view on that? Because if you're going to refactor your business, it takes a certain mindset, and there's some stuff that's not working on AI, but it probably will be, it's going to be human centric. What's your view on how to refactor it? I think there's going to be certain businesses that are totally reinvented by GPT, and open AI, and large language models. I think worst case though, there's going to be businesses that if they don't pursue that avenue, they're going to lose out on their users, their customers seeing a 20, 30, 40% improvement in productivity, and that alone is huge. Because we've all had dark in our products for years. We've all had chatbots, they're all terrible. And for enterprise software, I mean, look you can make the easy things easy, but in enterprise software, at some point you have to learn, you have to read the doc, you have to watch your video, and the documentation generally is terrible, because the writers don't understand the technology, and the engineers don't understand how to write. And so what I think is, I mean, why I'm pretty fired up about GPT is, it can break through, I think one of the big barriers that you have to adoption of enterprise technology is, how do you get the novice or easy users up to speed two, three times more quickly, right? For my business, we can't afford not to put people on that. Why, because we'll have a better product. Well, you think about how we get an answer to a question today. You don't ask Microsoft how to use Word or Excel, you ask Google, and then somebody else has answered the question for you. It's got to be a better way. Yeah. They're integrating into their Google and Microsoft. Okay, well they finally got it, right? But I'm just saying, you don't, okay. You don't search the platform for the answer. You search Google and then somebody else has the answer and then GPT is going to change that. I think we've got to get to a point where when we're releasing code, you maybe you have like an LLM ops team that's part of engineering, because if we have technically accurate information, we can train a model with that. And then we can let GPT write the beautiful English that is understandable by everyone. And to me, that has been the huge challenge in enterprise tech so far. And I feel like, okay, engineering, they should be good at explaining to people technically how to use a product. But now they don't have to think about how to explain it to a novice user because the model will do it for you. It's interesting. I know we've taught those many times in the queue over the years, data as code, infrastructure as code. If you look at GPT, a prompt engineering kind of discipline is you prompt the corpus. Now it moves to prompt ops. Yeah. The new thing I read a paper on is prompt tuning. So you have prompt ops, prompt engineering, which is just you really do prompt effectively and then operationalize that in the organization. And then now there's the tuning, which essentially having it take over on its own like self-healing networks used to be. So you have this discipline where if you do it right, the final stage is self-developing. And actually the human that is looking at the series of steps are, hey, was this helpful? Yes or no? A human is actually still useful. Right? Because a human is trying to perform a task, but they can also actually help in that training of the model. And so I see this as a great leg up for new users. Also, I mean, some of our users are very visual. They point and click around the UI. And there's been this big barrier between those people who like to point and click and those people who like to code. Well, if you can get GPT to generate the code, again, you're lowering the barrier of entry for the people who can perform a set of tasks in the past five minutes of code. This is where I like what you were saying earlier. I think this is where I see it compelling is if you go down a little bit on that progression, as you get smarter, the human trains the tuning side of it to learn on the data you're adding in while you're sleeping or while it's learning about like onboarding developers to sales development. You can see the go-to markets and all the business efficiencies that could be eliminated instantly. I mean, tech, I mean, I'm obviously, it's the only thing I know in my career. People say, you know, why don't you retire? I'm like, well, I don't know how to do anything else. You know? But what tech has delivered in the past 50 years is unbelievable productivity. And I think GPT, yes, it will absolutely replace certain industries and businesses. But worst case, I think you see in the next big breakthrough in productivity across the board. So you guys Silicon Valley folks, I got to ask you. I mean, what's the narrative is that growth capitals dried up, everything's moving into seed and early stage. You can start a company for 200,000 now instead of 2 million, right? I mean, what's happening in Silicon Valley? And then specifically, how is that affecting, is it affecting you guys? Yeah, I think, you know, after my old boss, Michael Dell used to say, you know, I like a good storm, it cleans the streets. And look, when capital is, when capital is free and easy, then you get a whole bunch of companies funded that shouldn't have been funded. You get a whole bunch of companies going public that shouldn't have gone public. And so what you see now is the storm kind of cleaning the streets. I mean, if we were here three years ago, we would have been talking about SPACs and people going public through SPACs. Not a lot of SPACs out there that are doing well. I mean, yeah, I think Silicon Valley too is also about the next big thing, you mentioned fashion. Bubbles tend to come out of that. But also opportunity recognition. Entrepreneurs have a fine sense of, you know, weirdness and brilliance that they hone in on opportunity recognition. And when you have in waves like this that come along that can change the game and clean the streets, clean the streets from all the garbage and trash and make it cleaner, but also actually pave new roads. And I think this market now is, it's a little bit bubbly, but there's definitely structural changes that are going to change how companies are built. I mean, remember the data center, you had to buy all the servers and then the cloud came along and that made it more efficient. We still had to pay for it and rent it. Now you got the, I can get to the product market fit faster, which is not saying just build a company, you can get to the validation faster, which came from the cloud benefit, but now you can do it faster and smaller and cheaper. Three people could do it, maybe. Okay, but back to Observe. So you came in in what, 2019, 18? Okay, so you were in that sort of spot where there's a lot of people that are going to get swept away in the street, but you guys are good? Yeah, by accident or design, we, I mean, Mike Spaza, who has been probably, not just the best investor, but one of the best champions for the company. After we did the series A, he said, hey look, you guys are quite some ways from market fit because what surprise does is the barrier to enter in this market is quite high. You need a lot of functionality before you can even get a meeting. It needs that look. If you do a series B now, you're going to get washed out, right? And this is, I mean, the age old Silicon Valley story of the founders that make it to market fit or beyond and they've got like half a point in the company's success at that point. And he said, look, why don't we run the company off of debt, off of a convertible note? So we've actually been running the company off debt. So we haven't done a series B yet. Cheap debt. Yeah, and so, I mean, I think one of the benefits is just having like a great investor. But now the product's ready for prime time, I do believe there is still like good money available for companies with good products and good business models. If we get through this year, exceed plan like we did last year, hopefully we'll do the series B next year. Yes, you do get dilution eventually. That debt does convert to equity, but we can do it at a time where we're going to have good revenue numbers, good customer counts, and we're going to have a business model that we can point to that the VCs can buy into, yeah. So that hopefully will be a growth capital round. Well, you guys worked hard. And again, what's great about your businesses, you're taking your time, and that's the plan. Yeah, the hard thing though. Well, I won't say that's the plan, but you're not in a hurry to force something. I think my expiser in particular at Sutter Hill is there's very few venture capitalists that say they take a long-term view and actually do. Most of them, as soon as you get your first couple of deals, they'll be like, hey, okay, if you get the number to this, then we'll fund you, otherwise we won't. And so they make the founders and the early employees do unnatural acts under the fear of them not getting funded. And so they actually do the exact opposite of what you should do because market fit, I didn't really fully understand it before I came to observe, is the hardest thing that I think you will ever do in your career as an early employee in a startup because everything, the details, they all matter to the nth degree and you doubt yourself and you don't know what the right answer is and you need investors that are going to give you time. Can you feel like you have product market fit now? I think we're very, very close. And when we were starting to ramp the team in anticipation, I mentioned Chi, who we hired from Stripe earlier, Chi's been amazing and for us, the final sort of like buffing off the rough edges is to get the product to spread more virally within accounts. We validated the economic model, we validated the value proposition, but can we get the product to spread more quickly within accounts? We can't have like a high touch enterprise sort of sales lift on every single deal that we do. You know? You don't get the sales productivity out of it. The innovation's messing, I mean, but getting it right matters. Yeah, Spicer has a great line, he says, you know, Jeremy, in one of my moments of despair, he said, Jeremy, the people who make it to the IPO, they're not the geniuses, they're the survivors. And that is the truest line ever spoken to because I think at the IPO, people look back romantically at history and it's brutal, you know. We like Mike Spicer, I mean, taking the long game and adding value, I mean, that's not so rare to find a VC's that do that. Incredibly rare. Jeremy, thanks for coming on theCUBE, great to see you, and then keep in touch, looking forward to more conversation and thanks for the scoop on your new threat intelligence model. Yeah, always a pleasure. We're going to write that up. Don't think we're going to miss that. Yeah, I'm the head of marketing, head of PR, I'm the head of this model. The PR department wasn't paying attention. Oh, you don't have a PR department yet. Okay, you're here to hear first on theCUBE. Good to see you, man. John Farmer, Dave Long, here in theCUBE. You're right back for more RSA Live coverage after this short break.