 Hello, welcome to this CUBE Conversation. I'm John for your host here in the CUBE in Palo Alto, California, featuring Cribble, a hot startup, taking over the enterprise. When it comes to data pipe lighting, and we have a CUBE alumni who's the co-founder and CEO, Clint Sharp. Clint, great to see you again. You've been on the CUBE. You were on in 2013. Great to see you. Congratulations on the cover that you co-founded and leading as the chief executive officer over $200 million in funding, doing really strong in the enterprise. Congratulations. Thanks for joining us. Hey, thanks, John. It's really great to be back. You know, remember our first conversation, the big data wave coming in Hadoop World 2010? Now the cloud comes in. And really the cloud native really takes data to a whole nother level. You're seeing the old data architectures being replaced with cloud scale. So the data landscape is interesting. You know, data as code, you're hearing that term. Data engineering teams are out there. Data is everywhere. It's now part of how developers and companies are getting value, whether it's real time or coming out of data lakes. Data is more pervasive than ever. Observability is a hot area. There's a zillion companies doing it. What are you guys doing? Where do you fit in the data landscape? Yeah, so what I say is that Cribble in our products, and we solved the problem for our customers of the fundamental tension between data growth and budget. And so if you look at IDC's data, data's growing at a 25% cager, you're going to have two and a half times the amount of data in five years that you have today. And I talked to a lot of CIOs. And the thing that I hear repeatedly is my budget is not growing at a 25% cager. So fundamentally, how do I resolve this tension? We sell very specifically into the observability and security markets. We sell to technology professionals who are operating observability and security platforms like Splunk or Elasticsearch or Datadog, ExoBeam. Like these types of platforms, they're moving protocols like Syslog, they're moving, they have lots of agents deployed on every endpoint. And they're trying to figure out how to get the right data to the right place. And fundamentally, control costs. And we do that through our product called Stream, which is what we call an observability pipeline. It allows you to take all this data, manipulate it in the stream and get it to the right place. And fundamentally be able to connect all those things that maybe weren't originally intended to be connected. So I want to get into that new architecture if you don't mind, but let me first ask you on the problem space that you're in. So cloud native obviously instrumenting everything is a key thing. You mentioned data, all these tools. Is the problem that there's been a sprawl of things being instrumented and they have to bring it together? Or is too costly to run all these point solutions and get it to work? What's the problem space that you're in? So I think customers have always been forced to make trade-offs, John. So the, hey, I have volumes and volumes and volumes of data that's relevant to securing my enterprise, that's relevant to observing and understanding the behavior of my applications, but there's never been an approach that allows me to really on-board all of that data. And so where we're coming at is giving them the tools to be able to filter out noise and waste, be able to aggregate this high fidelity telemetry data. There's a lot of growing changes. You talk about cloud native, but digital transformation, the pandemic itself and remote work, all these are driving significantly greater data volumes. And vendors, unsurprisingly, haven't really been all that aligned to giving customers the tools in order to reshape that data, to filter out noise and waste, because for many of them, they're incentivized to give as much data into their platform as possible, whether that's aligned to the customer's interests or not. And so we saw an opportunity to come out and fundamentally as a customer's first company, give them the tools that they need in order to take back control of their data. I remember those conversations even going back six years ago, the whole cloud scale, horizontally scalable applications, you start to see data now being stuck in the silos now to have high good data, you have to be observable, which means you gotta be addressable. So you now have to have a horizontal data plane, if you will. But then you get into the question of, okay, what data do I need at the right time? So is the data as code, data engineering, discipline changing? What new architectures are needed? What changes in the mind of the customer once they realize that they need this new way to pipe data and route data around or make it available for certain applications? What are the key new changes? Yeah, so I think one of the things that we've been seeing in addition to the advent of the observability pipeline that allows you to connect all the things is also the advent of an observability lake as well, which is allowing people to store massively greater quantities of data and also different types of data. So data that might not traditionally fit into a data warehouse or might not traditionally fit into a data lake architecture, things like deployment artifacts or things like packet captures. These are binary types of data that it's not designed to work in a database, but yet they want to be able to ask questions like, hey, during the log for shell vulnerability, what have all my deployment artifacts actually had log 4j in it in an affected version? These are hard questions to answer in today's enterprise or they might need to go back to full fidelity packet capture data to try to understand a malicious actor's movement throughout the enterprise. And we're not seeing, you know, we're seeing vendors who have great log indexing engines and great time series databases, but really what people are looking for is the ability to store massive quantities of data, five times, 10 times more data than they're storing today. And they're doing that in places like AWS S3 or in Azure Blob Storage. And we're just now starting to see the advent of technologies which can help them query that data and technologies that are generally more specifically focused at the type of persona that we sell to, which is a security professional or an IT professional who's trying to understand the behaviors of their applications. And we also find that, you know, general purpose data processing technologies are great for the enterprise, but they're not working for the people who are running the enterprise. And that's why you're starting to see the concepts like observability pipelines and observability lakes emerge because they're targeted at these people who have a very unique set of problems that are not being solved by the general purpose data processing engines. It's interesting as you see the evolution of more data volume, more data gravity, then you have these specialty things that need to be engineered for the business. So it sounds like observability lake and pipelining of the data, data pipelining or stream you call it, these are new things that they bolt into the architecture, right, because they have business reasons to do it. What's driving that? It sounds like security is one of them. Are there others that are driving this behavior? Yeah, I mean, it's the need to be able to observe applications and observe in user behavior at a fine-grained detail. So I mean, I often use examples of like bank teller applications or perhaps, you know, the app that you're using to, you know, I'm going to be flying in a couple of days. I'll be using their app to understand whether my flight's on time. Am I getting a good experience in that particular application? Answering the question of, is Clint getting a good experience requires massive quantities of data and your application and your service, you know, I'm going to sit there and look at, you know, American Airlines, which I'm flying on Thursday. I'm going to be judging them based off of my experience. I don't care what the average user's experience is. I care what my experience is. And if I called them up and I say, hey, and especially for the enterprise, usually this is much more for, you know, in-house applications and things like that. They call up their IT department and say, hey, this application is not working well. I don't know what's going on with it. And they can't answer the question of what was my individual experience. They're living with, you know, the data that they can afford to store today. And so I think that's why you're spending to see the advent of these new architectures because digital is so absolutely critical to every company's customer experience, that they're needing to be able to answer questions about an individual user's experience, which requires significantly greater volumes of data. And because of significantly greater volumes of data, that requires entirely new approaches to aggregating that data, bringing the data in and storing that data. Talk to me about the enabling customer choice when it comes to around controlling their data. You mentioned that before we came on camera that you guys are known for choice. How do you enable customer choice and control over their data? So I think one of the biggest problems I've seen in the industry over the last couple of decades is that vendors come to customers with hugely valuable products that make their lives better, but it also requires them to maintain a relationship with that vendor in order to be able to continue to ask questions of that data. And so customers don't get a lot of optionality in these relationships. They sign multi-year agreements. They look to try to start another, they want to go try out another vendor. They want to add new technologies into their stack. And in order to do that, they're often left with the choice of, well, do I roll out yet another agent? Do I go touch 10,000 computers or 100,000 computers in order to onboard this data? And what we have been able to offer them is the ability to reuse their existing deployed for prints of agents and their existing data collection technologies to be able to use multiple tools and use the right tool for the right job and really give them that choice and not only give them the choice once, but with the concepts of things like the observability lake and replay, they can go back in time and say, you know what? If I wanted to rehydrate all this data into a new tool, I'm no longer locked into the way one vendor stores this. I can store this data in open formats. And that's one of the coolest things about the observability lake concept is that customers are no longer locked into any particular vendor. The data is stored in open formats. And so that gives them the choice to be able to go back later and choose any vendor because they may want to do some AI or ML on that type of data and do some model training. They may want to be able to forward that data to a new cloud data warehouse or try a different vendor for log search or a different vendor for time series data. And we're really giving them the choice and the tools to do that in a way in which was simply not possible before. You know, you're bringing up a point as a big part of the upcoming AWS startup series data as code. The data engineering role has become so important and the word engineering is a key word in that, but there's not a lot of them, right? So like how many data engineers are there on the planet and hopefully more will come in from these great programs in computer science, but you got to engineer something, but you're talking about developing on data. You're talking about doing replays and rehydrating. This is developing. So data as code is now a reality. How do you see data as code evolving from your perspective? Because it implies DevOps. Infrastructure as code was DevOps. If data as code, then you got data ops. AI ops who's been around for a while. What is data as code? What does that mean to you, Clint? I think for our customers, one, it means a number of, I think sort of after effects that maybe they have not yet been considering. One, you mentioned, which is it's hard to acquire that talent. I think it is also increasingly more critical that people who were working in jobs that used to be purely operational are now being forced to learn, you know, developer-centric tooling, things like get things like CI CD pipelines. And that means that there's a lot of education that's going to have to happen because the vast majority of the people who have been doing things in the old way from the last 10 to 20 years, they're going to have to get retrained and retooled. And I think that, one, is that's a huge opportunity for people who have that skill set. And I think that they will find that their compensation will be directly correlated to their ability to have those types of skills. But it also represents a massive opportunity for people who can catch this wave and find themselves in a place where they're going to have a significantly better career and more options available to them. Yeah, and I think what we just said about your customer environment, having all these different things like Datadog and other agents, those people that rolled those out can still work there. They don't have to rip and replace and then get new training on the new multi-year enterprise service agreement that some other vendor will sell them. You come in and it sounds like it's saying, hey, stay as you are, use Cribble, we'll have some data engineering capabilities for you. Is that right? Is that- Yep, you got it. And I think one of the things that's a little bit different about our product and our market, John, from kind of general purpose data processing is, for our users, they're often responsible for many tools and data engineering is not their full-time job. It's actually something they just need to do now. And so we've really built a tool that's designed for your average security professional, your average IQ professional. Yes, we can utilize the same kind of data ops techniques that you've been talking about, CICD pipelines, get ops, that sort of stuff, but you don't have to. And if you're really just already familiar with administering a Datadog or a swamp, you can get started with our product really easily. And it is designed to be able to be approachable to anybody with that type of skill set. It's interesting when you're talking, you remind me of the big way that was coming. It's still here. Shift left meant security from the beginning. What do you do with data? Shift up, right, down? Like what does that mean? Because what you're getting at here is that if you're a developer, you have to deal with data, but you don't have to be a data engineer, but you can be, right? So we're getting in this new world. Security had that same problem, but a wait for that group to do things, it created tension on the CICD pipelining. So the developers who were building apps had to wait. Now you got shift left. What is data? What's the equivalent of the data version of shift left? Yeah, so we're actually doing this right now. We just announced a new product that we could go called Cribble Edge. And this is enabling us to move processing of this data rather than doing it centrally in the stream to actually push this processing out to the edge. And to utilize a lot of unused capacity that you're already paying AWS or paying Azure for or maybe in your own data center and utilize that capacity to do the processing rather than having to centralize and aggregate all of this data. So I think we're going to see a really interesting and left from our side is towards the origination point rather than anything else. And that allows us to really unlock a lot of unused capacity and continue to drive the kind of cost down to make more data addressable. Back to the original thing we talked about the tension between data growth. If we want to offer more capacity to people if we want to be able to answer more questions we need to be able to cost effectively query a lot more data. You guys had great success in the enterprise with what you've got going on. Obviously the funding is just the scoreboard for that. You got good growth. What are the use cases or what's the customer look like that's working for you where you're winning or maybe said differently? What pain points are out there that the customer might be feeling right now that Cribble could fit in and solve? How would you describe that ideal persona or environment or problem that the customer may have that say, man, Cribble's perfect fit? Yeah, this is a person who's working on tooling. So they administer a Splunk or an Elastic or a Datadog. They may be in a network operations center or a security operations center. They are struggling to get data into their tools. They're always at capacity. Their tools always at the red line. They really wish they could do more for the business. They're kind of tired of being this department of no where everybody comes to them and says, hey, can I get this data in? And they're like, I wish, but we're all out of capacity and we wish we could help you, but we frankly can't right now. We help them by routing that data to multiple locations. We help them control costs by eliminating noise and waste. And we've been very successful at that in logos like a Shutterfly or a Planking on End. But we've been very successful in the enterprise. That's my pick. And we continue to be successful with major logos inside of government, inside of banking, telco, et cetera. So basically it used to be the old hyperscalers with the ones with the data full problem. Now everyone's got their full data and they got to really expand capacity and have more agility and more engineering around contributions of the business. Sounds like that's what new guys are solving. Yep, and hopefully we help them do a little bit more with less. And I think that's a key problem for our enterprises is that there's always a limit on the number of human resources that they have available at their disposal, which is why we try to make the software as easy to use as possible and make it as widely applicable to those IT and security professionals who are kind of your run-of-the-mill tools administrator. Our product is very approachable for them. Clint, great to see you on the cube here. Thanks for coming on. Quick plug for the company. You guys looking for hiring. What's going on? Give a quick update. Take 30 seconds to give a plug. Yeah, absolutely. We are absolutely hiring. Cribble.io slash jobs. We need people in every function from sales to marketing to engineering to back office, GNA, HR, et cetera. So please check out our job site. If you're interested in learning more, you can go to Cribble.io. We've got some great online sandboxes there which will help you educate yourself on the product. Our documentation is freely available. You can sign up for up to a terabyte a day on our cloud. Go to Cribble.cloud and sign up free today. The product is easily accessible and if you'd like to speak with us, we'd love to have you in our community and you can join the community from Cribble.io as well. All right, Clint Sharps, co-founder and CEO of Cribble. Thanks for coming on the cube. Great to see you. I'm John Furrier, your host. Thanks for watching.