Mitchell Hashimoto on Company Origins, Remote Culture, Open Source, and Selling to Enterprises
May 12, 2019
Replicated's CTO, Marc Cambell discusses Mitchell's background, projects, and the growth of HashiCorp.
Founder & Co-CTO, HashiCorp
» Early career
I actually got started at as a sort of Rails dev consultant during that hype phase. I worked for a consultancy … and that's what sort of kick-started everything. I mean I was Dev, but I was always super attracted to Ops … we had Ops people so I always watched them and learned from them and it was really interesting.
» A remote-working culture
I wanted to build a remote company … the only people that aren't remote are Legal and Finance for … compliance [and] regulatory reasons … but everybody else is remote … that's like 275 remote roughly. … Part of the authentic experience [is] we have to from the top down and the bottom up create a culture and process that is remote-friendly.
I think it comes out of the natural open-source nature of what we do. … The way you work in open source is contributors who you've never met, anywhere in the world, submitting work asynchronously. And our best candidates to hire when we started the company were these people that were passionately contributing to the projects … with no financial incentive whatsoever.
I mean we've had at least five employees over the years quit—they would quit and say they love the company, love the products, loved my job [but] I need to sit next to someone. … They felt lonely. … We've tried stuff to counteract that—we've tried out sessions where there's open rooms that people could just go meet random people … it's like chatroulette though. [But] it filters certain types of people out.
» Where the ideas came from
Armon's my co-founder. He's a genius—bona fide. He graduated college two years early, went to college two years early—just super super smart. … We met on a research project … at University of Washington in Seattle. … In 2008, Seattle is … Microsoft Redmond, Amazon across the lake, and Google which is pretty close. … And they were all dabbling in this cloud idea. … So a lot of the research was around this.
Microsoft gave us servers in a box—it was the data center in a container but like a physical container and Google gave us some computers, Amazon gave us some EC2 because EC2 was out, and a bunch of universities donated really crappy computers to us. My job on that research project was to figure out how to manage all those servers in an automated fashion, because we had like 10 students on the project and so I had to figure out how to wipe them, how to install software, how to deploy software…
And they were running multiple operating systems so we were we had another team member team actually that was studying Rails and so we were studying Linux containers … we're studying BSD, … we're doing custom OS X kernels … extensions to do stuff. … Being at a university is cool, because we had source access to Mac OS, Windows and of course Linux open source, so we got to see all this and work with their engineers. That's what's cool about academia.
That's what gave me and Armon … the idea for Terraform and Consul. … It was all written down during those years in college. … And he walked away from that super interested in the space … and so he emailed me out of the blue and wanted to talk about it and we quickly started having breakfast every single day together, we started coding together. … Armon's way academically smarter than me—I was not a good student—so he absorbed all the theory more than I did and so he brought a whole new layer of rigor and cool ideas and foundational stuff to the solutions
» Other topics:
- Open-source users expect free.
- How do we sell to the fortune 500?
- The importance of having investors and advisors.
- There was one year—one whole year—he went without coding, and that that was the worst year.