 Hello everyone, welcome to this special CUBE conversation. We are here in Palo Alto, California at the CUBE's Silicon Valley headquarters. I'm John Furrier, the special guest here. Mitchell Hashimoto, 25 year old CEO of Hashacorp. Welcome to the studio here for the CUBE conversation. Thanks for having me. So, software's eating the world as Mark Andreessen said. But we always talk about that on the CUBE. It's kind of a punchline. It's become kind of cliche at this point. But one of the things that we really get psyched about is innovation around how fast the software business can move and you've been a great example of that. Your company has been shipping code heavily for a couple years now with Vagrant and then now commercializing that and helping customers. So, super excited congratulations. Thanks for spending the time with us today. First question, what's going on with your world right now? Vagrant, talk about the Vagrant dynamic, what it is, how it got started and then how it's maturing right now because with Docker containers, everyone's seeing the cloud right now mainstream. The cloud is here and DevOps has been around. We've been part of the cloud of Roddy talking about, oh, cloud's great. You abstract away complexities of infrastructure. Dev and ops, great. But now with the Docker container movement and some of the stuff that you're doing, cloud computing is mainstream. It's here to stay. The numbers are being posted by Amazon Oracle ship. Some numbers this week. Cloud computing, it's a done deal to score board. People are making money. Real value is being created. So, take us through the Vagrant story and how it matured out. Sure. So, Vagrant is our oldest software. It's six years old. I mean, it still is a developer focused tool for building development environments and it's really what kick started for us. What kick started this whole thing and it has the probably biggest user base, broadest user base of everything we do and it's any developer, it's useful for them and in relation to containers, I think a big shift has been that Vagrant, given that it is six years old, creates development environments primarily with virtual machines and as we're moving more towards containers, people have a bigger interest in containers. The biggest value still of Vagrant though is that it gives you the same development environment whether on Mac, Windows or Linux, developing in Mac, Windows or Linux. So, any sort of part of that matrix in there, like if you're on Windows and building something that runs on Linux, it'll set that up. If you're on Linux and building something that runs on Windows, it sets that up and that's still a huge value. So, there's containers which have more reportability problems just because you still need virtual machines on things like Windows to run Linux containers and so on. Like, Windows has like Hyper-V containers coming but still only runs Windows stuff. So, Vagrant is still a unifying tool in terms of getting a consistent environment across all the platforms. Where's unification coming in because this is a real major point because with cloud computing, obviously now mainstream, I quoted some of the numbers from the big guys and it's in the billions of Amazon, hundreds of millions of quarter and who knows what other numbers are. It's massive, there's pressure. If the startups don't move fast enough, these communities could crumble or have to rally around either one standard or multiple standards. So, there's a lot of pressure. So, standardization and unification is the topic that everyone wants to talk about. What is the dynamic going on? What do you guys do differently? Is it because of the VMwares? VMs are the VMwares, VMs? Or is it because that's more mature than say the container market? Container market right now seems to be great. People love it, it's intoxicating. Hey, rah, rah. But there's not a lot of native cloud apps, if you will, shipping right now as much as there is virtual machines. Right. I think in terms of, it's more unification than standardization, it's sort of two separate things. Like the AWS and Google and Microsoft all have their own sort of APIs, their own things. So, instead of viewing it as a standardization, like I think there's open stack as well but that's sort of, it's also its own thing. So, that would be like number four. So, there's all these different things. I think unification is the key. So, for that it's more about our other tools. So, we have other tools that are more production data center focused and for them what we've done is tried to just hide that layer for you. So, Terraform is a good example. Terraform is a tool that builds infrastructure and it could do it on Google, Google Cloud, Azure, AWS, VMWare, open stack, all these things. And so what we do is VMWare is the layer that talks down to these different APIs but we sort of expose the same configuration interface on top of it. So, I think that's more important and containers are a very similar thing is that you like, you package something up in a container as long as you have Linux and it's running somewhere for now that it runs whether on any cloud you have. So, I think it's just a... So, there's no issue with provisioning tools like Puppeter Chef, you guys work with pretty much. Yeah, for us we don't build config management tools so we try to iterate wherever possible with them. Absolutely. Talk about the demand that you guys are having in the marketplace. Where's the action? What's the top conversations that you're involved in with developers? Developers are the canary in the coal mine. They do telegraph kind of where the market's going. What's going on with the developers in your world? What are they solving for use cases? What's the problem cases? What are they going after? From the developer perspective, I think the biggest problem still is just more efficiently sort of getting it out to production. It is speeding up still the process of starting something new or working on something legacy in either way and getting it out to production. I think that's the biggest challenge from the developer point of view and operators still though are, that's the same thing I would say, but they're viewing it from a different perspective. Developers are like, I have this thing, I want it to go out and operators more like see that, oh I see you have that thing but I don't know how to get it out there safely quickly. So I think those are the two sides that I think the goals are aligned though that people want things to move faster. As a production on premise or in cloud or both? Both, yeah, both. What is the number one use case you see people working on in terms of their outcome solutions? So when they actually are building the products, what do you see as the use cases for the developers and the product? Is it like test and dev? Are you seeing more workloads that are more tied to say financial services? What kind of color can you share with what's being developed out there with you guys? Use cases in terms of what? The kinds of apps that they're building. Oh, kinds of? I mean, all of them. I think that there's just a huge boom right now in everything. I think we're seeing a lot of big companies even like rewriting very legacy components because of this shift to cloud things. If there's a potential huge cost saving, actually better security running in the cloud than your own data center, things like that. So there's a big motivating factor to get into the clouds. I think across the board we're seeing everything. What's your take on DevOps as it evolves to mainstream? Because this is a fun conversation always. This is like, how do you hire a DevOps person? What is the, is there like a profile because it's a unique individual? Now the young guys like you guys are born with DevOps. So it's kind of a DevOps native mindset. But as it goes mainstream, a lot of these enterprise have guys that are like my age or kind of in their 30s who are used to load Linux on machines and load patches. So like there's a whole new world coming into that's DevOps is being introduced to because of the relevancy of it. What are some of the things that you see in that culture right now? If you had to explain that to someone in an enterprise like, hey, you know, Mitchell tell me about how do I do DevOps? Or who do I hire? What is the ethos? Yeah, I think if you ask a thousand people you get a thousand different answers. But sort of the way I like to break down DevOps is it's just about moving faster from development to production. So it's like, do you wanna get to production faster? But while still being safe, then DevOps is what you need. And DevOps really is about automating things. So automating process, getting process overhead out of the way. And when you automate things, developers could take part in more things because there's not a knowledge bottleneck. There's not a person model like that sort of thing. So if someone were to ask me how do I do DevOps, I would just say look at, I like to say get a whiteboard and write down when you start working on a developer writes a line of code to when that thing's in production and a problem happens and how it gets respond to. Write down the whole every single step that happens. And it'll be a lot. And just write down every single step that happens and just figure out how to start automating those steps away like one at a time. It doesn't need to be, you don't need to hire like a DevOps team that just shifts the entire thing in an atomic switch. I think that could be disastrous. I think it's easier to just slowly just do a piece at a time. Yeah, that's a cultural thing too. It's not so much they press a button and instantly have a DevOps organization take some learning and some culture, right? Yeah, definitely. And I think the challenge is primarily cultural, I think for bigger organizations or legacy organizations, but there's also technical challenges but I think those are disappearing every day. I think the technical challenges are mostly like Chef and Puppet or the original DevOps, what grew with DevOps, but now there's these other tools and even like Docker, it's like, well, I have a Windows application, like how do I automate the Windows application? And for a long time, the tools were lacking in terms of Windows, so you're technically hamstrung as well. And now I think those are disappearing every day. Break down the top three things that you guys unify that you help solve. So the top three things is development environments, obviously the vagrant, but automated data center resource creations, creating servers, creating load balancers, network things, things like that. And then the other thing is probably distributed system or microservice orchestration is the third thing that we do. And those are really widely varying topics but we sort of do a lot of different things. What's your philosophy on data? Data layers, data lakes, I call them data oceans, but data is a big part of the intelligence of, you know, real time, if you're going to go dev ops, you're going to be agile, you need data. So is there a philosophy that you guys have with your architecture around how you view data, where it's stored, how it's accessed, and so on? Yeah, I mean data is a very big limiting factor. It's hard to beat the speed of light. I mean, people want, I think the ultimate dream is to be able to just run the database or anything anywhere, but at the end of the day, you have terabytes of data somewhere and you just can't transfer terabytes of data faster than the speed of light. So you're going to be stuck with that limitation forever. So I think that rather than fighting that, we just need smarter tools that focus on putting the data, heavy workloads, map-reduced jobs, databases, stateful services, things like that, near the data. Like we should stop trying to make the data invisible in terms of it moving around, and we should just make it invisible by moving the compute to the data, right? Exactly. That's a great philosophy. You tell you get the dev ops, mine cranking right there, I love that. But that's a lot easier because you have a lot more resources. I mean, in theory, it's limitless, I mean, quotes. Yeah. It's getting smaller and cheaper and faster. Yeah. Okay, so I got to ask you about the company. Tell us what's going on and how do you hire people? How do you hire the people in your culture, whether it's dev ops engineers, designers and or staff? Sure, so HashiCorps, I guess, as a company has only been around for a few years, but as a technical entity has been around for probably six. And what we've been up to for the past few years has really been shipping a lot of software and actually moving from development more and more into production. Actually, like four out of six of our tools now are very production focused. And most of the conversations I have on a day-to-day basis are actually now with operators, more so than developers. And I think that's just what we're trying to do is really, we knew we couldn't solve the problem of getting from development to production without having a stronger influence on the operator side of things. So we've been focusing on that. So a lot of our newer stuff, a lot of our fastest growing stuff is data center related, creating data centers, like I said, service discovery, configuration, cloud security, stuff like that. And that most developers actually, it's funny when I walk into a conference, you either know me from Vagrant nowadays or you don't actually maybe even know Vagrant at all and you know me from some other data center thing. So it's kind of a funny breakdown. But you guys look at it from a holistic standpoint, tools, built some great tools and then the totality of the tools makes up the market basket of products, right? Or is there a platform as well underneath that? No, so yeah, our vision is to create a bunch of, so we have six open source projects down, create a bunch of very tactical targeted tools that solve one thing really well. But the holistic vision that we have is that you'll be able to combine these tools to get the end-to-end developer to production. Yeah, kind of like the Lego blocks, as we always say, it's been overused, but still relevant, kind of services-based. What's your take on Silicon Valley now when you're out here in the city? What's the culture like and how do you look at the current market? A lot of people say we're in a bubble, but we always counter that with a lot of innovation going on up and down the stack and you see horizontally scalable proven itself. Now the vertically integrated from chip to edge, you're seeing optimized software. So you have a lot of different competing approaches, a lot of great young companies like your company growing. What's your philosophy of what's it like in Silicon Valley and the Bay area here right now? It's, there's a lot going on. Like you said, whether it's a bubble or not, I think that there's a lot of money flowing and whether that's positive or negative. One of the positive things about it is it does fuel innovation in every sort of category. So seeing a lot of new things really fast and yeah, I think it's a good thing right now. And I think the big thing, at least in our space, is that a lot of startup companies like ourselves and Docker, CoreOS, people like that are effectively changing the way enterprises are doing operations. So it's sort of a, there's sort of a big enterprise trust that's now coming to these newer companies. Like none of the companies I just named is more than three years old if you don't count Dock Club. But yeah, none of them are really over three years old. So it's neat to be able to get that sort of trust and that sort of influence. And I think I see it from that perspective, but there's also a lot of the network companies, storage companies that are popping up that are challenging enterprises right now. But a lot of the stuff that's coming out open source, like you guys are a great example, is a collection of years of work. So it's not like just add water and some sugar and you have an instant like home run. I mean, there's some cases out there where you see massive funding and it gets a little boost. But a lot of the heavy lifting is being done from small teams that have kind of worked together in open source for a few years and then pop on something. In your case, you guys kind of had a nice trajectory off the work you do with Vagrant. And look at what Solomon did at Docker. I mean, he had, it was not a failure. Or anyway, just a small little pivot when they saw the movement on containers. So again, it's being in the game and then taking advantage of whatever's available. Do you see it that way too? And what's your advice to folks out there that are in open source that are trying to figure out how to start a business and how to get going? Yeah, I think that's a very good point that there's, and that's the same way with HatchCorp too, is there's a three year old company with six year old technology. So it's a good point. I think my advice if you're working in open source is make sure you're just, you're solving a problem that people have. There's a lot of people in open source that build something because it's cool, which is totally fine. Building a business on something that you just did because it's cool is gonna be difficult. This hobby versus kind of building a problem that you go, hey, actually make money out of it and repeating it and actually selling it. It's two different things, right? Yeah, so like Docker, for example, solved the huge pain point, which is just deploying something in a standardized way, like format, right? And that was a huge pain point in terms of the speed of getting things out. So it's just sort of identifying what are the pain points of things happening. And you don't need to be in these big companies to see that. Like when I built Vagrant, I worked for a consultancy and the pain point that I was seeing directly at the consultancy, but all the companies I was working with was that developers couldn't get up and running quickly. So I built Vagrant and that was sort of, it was for me and it was fun, but it was also solving a problem. And I think that having those things together made it eventually to help with an instant success. Well, it keeps you focused and then you have a checkpoint against whether you're solving the problem. That's a good way to do it. Okay, so final question is, how do you guys hire? What's your philosophy? How do you hire DevOps guys in your company? You'd whiteboard them, put a little quiz, you pick them out, what do you do? How do you vet out the quality candidates? Because you don't have massive numbers. So it's also your team, your approach. Cher, your insight. The top thing we look for is passion. This is a not super sexy sort of industry. I think it is, but like a lot of most people don't. So it's hiring the people that actually are passionate about this industry, which really sets them apart. So the person who really cares about automating things and cares about making sure servers don't go down and things like that. So the way we hire is look for people who have done ops. I think that's a very important point. You need to understand the pain in order to create the solution. So look for people that have done ops. We look for an overlap and people that like enjoy open source and see if they're passionate about, have the drive in order to like automate things. And that's what we look for. And you guys are in San Francisco, right? Yeah, we're based right in San Francisco. One office? Do you have anyone remote league to virtual teams? Yeah, actually, over half our team is remote. So we've hired, a lot of our early hires were from our open source community. And so we knew their work was great. We knew their personalities were great. And they're just sort of all around the US. That sounds like another thing too. Just get in, start contributing. If people can see you, you throw the ball around kind of the field, if you will. Yeah, it's a great way to get a job with a company that does open source. I think that everyone doesn't have the privilege of being able to have time to work on open source. But if you do, then it definitely catches attention. All right, so shoot the arrow forward five years. What's happening in the market? What's going to be the world like in five years? If this trajectory continues, depending on how the slope is, but DevOps and automation certainly relevant seeing it at the data center on premise, hybrid cloud with the big guys and then maybe pure cloud. What's our world look like in your mind? I think in five years, everyone's on a cloud of some sort. And I think that for the most part, data centers will be almost completely self running. And they'll, the pain point of data centers is going to be marginalized to a very small amount relative to what it is right now. And so developers will just be able to deploy things very quickly. Operators will be able to monitor them very easily. I think everything in five years is a long time. I think everything will be very seamless by then. And your goal for Hashacorp is to do one, go public, have a lifestyle business, produce great cash flow. I mean, what's the vision for the company? Well, the vision for the company is to be managing all those data centers in five years. Yeah, hopefully not answering the question of how you begin to sell or go public. But you want to build a big company. That's the goal, right? Yep, definitely. Okay, great. And thanks for spending the time to come in to the CUBE office here at Palo Alto. Appreciate it. Mitchell Hashimoto here with Hashacorp, big rent, great software. Check out the site, hashacorp.com and stay tuned for more action from SiliconANGLE.tv, it's the CUBE conversation. Thanks for watching.