Will JavaScript type annotations kill TypeScript?
The creators of Svelte and Turbo 8 both dropped TS recently saying that "it's not worth it".
Yes: If JavaScript gets type annotations then there's no reason for TypeScript to exist.
No: TypeScript remains the best language for structuring large enterprise applications.
TBD: The existing user base and its corpensource owner means that TypeScript isn’t likely to reach EOL without a putting up a fight.
I hope they both die. I mean, if you really need strong types in the browser then you could leverage WASM and use a real programming language.
I don’t know and I don’t care.
Open Source / Tech Life

HashiCorp’s Mitchell Hashimoto on When to Step Down to the Job You Love

A Q&A with HashiCorp founder Mitchell Hashimoto, on his decision to step down from his executive position within the company.
Jul 28th, 2021 12:00pm by
Featued image for: HashiCorp’s Mitchell Hashimoto on When to Step Down to the Job You Love

Mitchell Hashimoto, co-founder and namesake of HashiCorp, which offers an open source, enterprise infrastructure suite, announced last week that he would be stepping down from the c-suite to participate only as an individual contributor. Unlike other open source founders as of late, this was not because of scandal, abuse or community neglect. Nor was it to move to another high-profile role at another unicorn company. This was more about personal preference and aptitudes. And while it seemed quite sudden to the doting HashiCorp community, it was two years in the making.

The New Stack sat down with Mitchell Hashimoto to learn about his decade of leadership at HashiCorp, how to build an open source business, and why he made the decision first to step down first as CEO and then as Chief Technology Officer, all for a love of engineering.

This interview has been edited for brevity and clarity.

Let’s talk about your tech journey. What led you to start HashiCorp?

I went to University of Washington. I was pretty much equidistant from Google, Amazon and Microsoft. Being an undergrad, their presence is everywhere. And one of the things they were pushing at that time was this idea of “programmable data centers” — you know, “cloud” wasn’t a common word used then.

I hopped on a research project, and my job as an undergraduate was to figure out how to deploy applications to multiple programmable data centers. Being a university research project, we got a bunch of resources and computers donated to us. So I had to figure out how to deploy to Amazon where we had some credits, I had to figure out how to deploy to this precursor to Azure that was given to us, and to what you would call on-prem today, which is sort of just donated hardware that was in our hardware lab on campus. 

I failed abysmally on that research project.

But someone who was on that research project with me was Armon [Dadgar, co-founder HashiCorp.] We talked about what went wrong. I said, “I feel like there was no way that I could have succeeded because I’m missing all these tools that I need.” And some of those things I wrote down ended up being the HashiCorp tools. 

Because I always used open source, that wasn’t even a choice. In my mind, it was just the default. 

When did it move from a passion project to an actual business?

It never even once crossed my mind that I could possibly start a business around this. Then I moved to San Francisco and started immersing myself working for another startup, in that culture, and meeting people and venture capitalists. There were a lot of people telling me they were aware of the project, they used the project, that we should start a company around this. But no one knew exactly how the company would work, who you would sell to, how you would make money off this open source. And it was also the thing I loved working on.

I was sort of forced to start a company because I would go to work nine to five, go eat, go to the gym, dinner, be home by 7:30 p.m., and then I would work on this stuff until one or two in the morning. It just reached a point where I had to make a choice of one or the other. And I chose the open source projects.

One of the key challenges of open source is that it’s not naturally a business model. How did you decide on the freemium model?

Open source is not a business model. They’re separate dimensions to me on the graph. You could have a really good business model and be open source — I think we’re hopefully an example of that — but you could have a really good business model and not be open source. So when people say ‘I’m starting a business in open source,” I always ask: Why? What are you trying to do? Are you just trying to share technology? That’s a good reason. Are you trying to build a community of contributors? That’s another good reason.

I think we’ve gotten better at being HashiCorp the business, and that was the wild card early on. I think every investor that was investing in us knew: “These people can build software, that’s not a concern, but can they actually build a business around it?”

This is the first company real company I ever really started, and so a lot of the ignorance associated with that comes with it. Early on, Armon and I thought we’ll create something that you swipe a credit card, and anyone that has money, whether you’re an individual or a Fortune 500, could just do it. We didn’t realize that selling to a Fortune 500 is a totally different thing, that features that a Fortune 500 wants versus an individual are a totally different thing.

At a certain point, we realized that all our big users were begging us for certain features, and they were also begging to pay us. I learned that a company would rather pay a vendor, not just for support but to ensure they’re sustainable.

Three years into the business or so, we decided that we would focus on enterprise as a business, and we would create an open core base model of an enterprise version — the open source version. And we would focus very enterprise-y features into the paid version that hopefully wouldn’t upset our open source user base. And I think we ended up doing that pretty well.

How has your role at HashiCorp changed? Talk about your decision in 2016 to step down from CEO to co-CTO.

It was about four years into the company. We had raised a couple of rounds and were about to raise our third. I had learned in that time a lot more about what the CEO role entails. I think, if you watch TV or pop culture, a CEO felt to me like a pretty nebulous role, and I was learning what the role actually entailed. I realized this was a lot of executive management, executive building — we didn’t have very many executives at the time, so hiring. Financial management, both in terms of what we’re spending, of course, currently, but what we’re raising, when we should raise, how much we should raise, and what terms we should raise on.

Then there is the startup aspect, where a CEO does a lot of things that, in the future, you would delegate — messaging for marketing, initiate sales programs, things like that were all things I was either directly responsible for or helping with very much. 

There is a lot of external peer pressure of what you’re expected to do, like everyone else does this so you should do this too. I think it’s OK to do what makes you happy.

And I just realized that this was not something I particularly enjoyed. And also it’s not something I’m naturally good at. It’s something I felt I could maybe learn, but, like anybody else, in learning, I’d make a bunch of mistakes. Was this the best use of my time for the company to learn how to be a CEO? Armon and I thought we needed to focus on product engineering leadership, and that’s when we started the search [for a new CEO.]

Why did you make the bold decision this July to step down as CTO at just 32 years old?

As the company enters this later stage startup environment, HashiCorp is over 1,400 employees. I started realizing what a bigger company CTO is actually responsible for. And there’s also this aspect, there’s certain founder responsibilities that you can’t get rid of. Despite being CTO, because of the co-founder executive role, it still involved customer meetings, leadership planning, management around go-to-market strategies, customer success building and organization building, outside of engineering and product. 

It was taking more and more time away from engineering. And it should. I think an active executive CTO shouldn’t be hands on keyboard. I didn’t mind all this. For example, I genuinely enjoy talking to our customers, but it’s so time-consuming. If I had to choose one or the other, I liked engineering more. 

About two years ago, I had a conversation with David [McJannet, HashiCorp CEO] and Armon. Do you think there’s a path where I could become an individual contributor again? That’s where my number one passion lies. 

And we charted out this path. 

How will your new role be defined?

I feel like we are 80% through the transition already.

Since we’ve been doing this transition for two years, the fact that I’ve been more engaged and more excited is a sign that this was the right choice for me. I don’t think I’ll be working less, but I will say I will be working healthier hours. It wasn’t expected of me, but I always just forced in extra hours to do more engineering work because I really wanted to program and do that individual contributor work. I was sometimes a 40-hour-a-week executive and I would do another ten hours of programming because I wanted to.

As I stepped down from CTO and now am just the co-founder, I’m still going to be part of strategic meetings with regards to product engineering, but I am now yielding responsibility for the go-to-market side or anything non-product engineering-related.

I’m going to continue to be part of our conference keynote planning. So that is a bit of marketing, but our audience, they’re all engineers, mostly, so that’s usually highly technical.

Lastly, I’m going to keep meeting with Dave and Armon weekly, but it’s going to be more of an advisory relationship at that point and less of a final decision-maker.

I’m not part of executive weekly meetings. I’m not part of executive off-sites. I’m not part of the board anymore. And that’s where a lot of these decisions would happen. I’m sort of explicitly out of the information flow directly, and I’m more of an as-needed type of role.

What will you work on now? Who will decide that?

Yes, there are certain cultural power dynamics that can be scary. But we’ve had this figured out over the last two years. I would work through our vice president of engineering and talking to a team and figuring out where is a place where I could provide value. Either helping bootstrap a new team and get a new vision laid out or helping build a particularly confusing feature. My whole thing is I understand I’m a founder, I understand I was CTO, but what’s your process? If you’re the engineering manager, you’re the manager of this team. Tell me how you do tickets, how you do process, how you do stand-ups and performance reviews.

I’m really excited about having the freedom to spend a little time to just mess around. Maybe having 30 hours a week of high-impact work and then ten hours of ‘Let’s just try something.’

I’ve been full-time on the Waypoint team for about a year now and I’m going to continue working with them as that team matures and grows. At some point, I’ll step off that project.

Community is inherent to open source. How has the HashiCorp community reacted to your stepping down?

It appears 100% positive. I haven’t really received any negative feedback whatsoever. 

It was really exciting because there was always something in the back of my mind, because it’s not a normal path to take. But I think, if anything, all the support from the community, from the internal reaction as the employees found out, emails I’ve gotten, and phone calls we’ve gotten from our VCs and investors involved in the company and board members, it’s just reaffirmed that this is a really good decision. Good for the company. Good for me. 

This was an unusual path to take, choosing to essentially demote yourself. What should other tech organizations learn from this move?

I think it’s very important for tech companies to have a good IC [individual contributor] track, as well as a good management track, including senior individual contributor roles. At HashiCorp, our most senior role currently given our size is a principal engineer. Bigger companies, like Google and so on, will have staff engineers, fellows and distinguished engineers. That is a really big source of innovation for a company. 

The other side of it is recognizing that you should balance personal happiness with what others expect of you. There is a lot of external peer pressure of what you’re expected to do, like everyone else does this so you should do this too. I think it’s OK to do what makes you happy. But it was balanced. I did what was good for me, but I did it over a two-year period, so we could figure out what the right transition process was, so that it’s also good for the company. 

You are in a position of privilege to be able to take this step down. How could other people take similar steps? Are you really assuming risks financially or career-wise?

This is why I try to limit the generalization for this too much because I would put it exactly that way, I am privileged to be able to make this choice. For me, there is some risk for sure. I’m no longer on the board. I’m no longer on the e-team. So my voice is as-needed. If there’s something I don’t agree with, it’s not something I could force in there. I think that’s the risk to me.

There are compensation changes associated with this change, but, at the same time, it’s not a big risk to me because, being a founder of a company, most of my financial outlook is tied into stock. 

Group Created with Sketch.
TNS owner Insight Partners is an investor in: The New Stack.
THE NEW STACK UPDATE A newsletter digest of the week’s most important stories & analyses.