What is the HashiCorp origin story?
Jan 16, 2019
Hear the story of how HashiCorp began from co-founders Mitchell Hashimoto and Armon Dadgar.
- Armon DadgarFounder & Co-CTO, HashiCorp
- Mitchell HashimotoCo-Founder & CTO, HashiCorp
Armon: So, Mitchell, I know a thing that I get asked all the time is, how did HashiCorp come to be? What was the origin story? So I thought maybe it would be interesting if you wanted to share a little bit about how it all came to be.
Mitchell: Sure. So both Armon and I went to the University of Washington in Seattle. And, I think, we were both on a project also called Seattle. But being in that city really gave us a unique opportunity, because we were very closely surrounded by Amazon, Microsoft and Google.
And so we had access with our undergraduate research project that we were on together, to basically program against early versions of "the cloud" for these three companies. And this was long before the greater industry was using cloud.
It was around 2008, so AWS was brand new, and Google and Microsoft hadn't launched any official version of their cloud, even though they were giving us resources. And Armon was more or less my boss on this project. And my job was to come up with a multi-cloud deployment solution because we were a university project—low budget, grant based. So we used whatever computing resources were given to us.
And maybe we had two servers here and two servers over here, and we wanted to be able to deploy all of them. So that was my job. And I like to say I failed pretty spectacularly. I couldn't get it to work.
And when we were retro-ing—again, Armon being my boss—what went wrong, we discovered a bunch of problems that ended up becoming the groundwork years later for HashiCorp.
Armon: Yeah, and I know in some sense it sparked a little bit of our entrepreneurial spirit. I still saved the email I sent Mitchell. It's maybe 11 pm on a Thursday or something, being like, "Hey, I don't know what you're doing, but maybe we should start a company."
And I think, you know, instantly got a reply back from Mitchell. We went back and forth and agreed, "Okay, tomorrow morning we meet and start kicking this off."
Mitchell: I think I responded that night.
Armon: I think you did, yeah, just a few minutes later.
Armon: And it was funny. It was first of many failed attempts, I think. At the time, that project was called Amped, after the first initials of all the people we gathered together.
And it was a learning experience as we went through bad idea after bad idea. Ultimately, we ended up abandoning it and both moved to San Francisco and working at the same mobile ad company, which I think—funnily enough—we found that years later many of those cloud problems were still there.
Mitchell: Yeah, and like I said, the years we were hitting the problems initially were like 2008. And so clouds were definitely not used by the enterprise let alone startups at a very basic level. And so we didn't view those problems that we hit as a market opportunity.
And so we kind of abandoned it for a while. And it was only moving back to the Bay area and being surrounded by a bunch of tech companies that we realized like, "Wow, a lot of these people are hitting the problems that we hit like four or five years ago."
I wrote down the notes for that retro in a moleskin notebook, and I still have it. And it was like four years later that I opened that notebook and I was like, "Yeah, we wrote all these down. We should try to solve these"
Armon: And then I think you left first—I think it was November 2012—left the ad company, started HashiCorp, and I think our discussion continued as to, what's the ambition? What's the vision here in terms of what HashiCorp wants to do? And I think ultimately we agreed that all these problems are still unsolved.
Everyone seems to be re-solving provisioning, and service discovery, and security, and application deployment. And we said, "Why not?"
Mitchell: Why not?
Armon: "Let's take a stab at doing it once."
Mitchell: Yeah, why not us, I guess.