AMA with Eric Johnson, CTO (Public Stream)

Hi, I'm Eric Johnson, CTO. This is Monday, August 23rd, and I'm doing an AMA here at GitLab. I don't see anything in the docs, so if someone wants to feel free to shout out their first question, we can just verbalize it.

Yeah, I don't know anything about IBM DOORS, so I can't do a diff there. And for the Manage team in particular, I know they're working almost exclusively on tightening up security at the moment, so there isn't a lot of velocity coming there. But Plan is broader than that. I'd like to see us use more of our Plan stuff internally. I know there's a series of issues that Kristin, the PM, and Christie, our VP of UX, have identified, mostly around entity relationships, to allow things to work a little bit better for internal usage. And then I think we can use more Plan up to it, including the portfolio management stuff. I'd eventually love to see a long-term Gantt chart hydrated with our major initiatives like SaaS first and a lot of the scalability work that we're doing over time, and then progress against each one of those things, kind of the 40,000-foot view. As an executive stakeholder, I feel like we sometimes have to build those ourselves if we want them, using spreadsheets and whatnot. I think the goal should be that we're using our own app and using those feature sets.

That was her job. Anyone want to go next?

Yeah, so to step back on what engineering allocation is, we've documented in our handbook for specific teams percentages of any given release where the engineering manager or engineering management is going to prioritize what we do, mostly for the sake of reliability. That's the big one. And because we've got some debt to pay off, some of those percentages have been dialed up quite high. So what that means is the back-end engineer persona are working almost exclusively in the short term on availability work. And then it begs your question: we've got front-end folks, UX, and PM. Those are, I think, the three other personas that, if we're not very deliberate, might be kind of left out in the cold. And of course, we don't want anyone to be idle. Luckily for us, there's plenty of stuff to do. I know you and Anup have been working on this. We've got a component library called Pajamas, and there's a bunch of unconverted UI components that front-end could work on. There's a whole collection of UX debt issues that could be executed. There's a lot of forward-looking research that could be done. And of course, the PMs, which is not my area, Scott and Anub's area, have a bunch of forward-looking work that they do, could be competitive analysis, proof of concepts, or research collaborating with UX research, other sorts of things. So I encourage every leader in that area to take a look at the next few releases and make sure they're being planful and setting people's expectations. I think there's plenty to do; no one's at risk of being idle.

Who's next?

Yeah, I think it's natural to take your local view, like for us the US view, and so there's complexity even how it's going within my own country. But we're such a global company, we've got to be mindful that there's other folks. We're talking in the US about what percentage of people are choosing to get vaccinated, and meanwhile we've got many team members in countries that don't have access to a vaccine, let alone choice for that. So I think we need to be mindful when we make global decisions, especially about travel and things like that. Obviously, we didn't do an employee conference this year, and that was one of the big inputs into it. Locally to the US, my wife and I have been vaccinated since late April, but we've got small children, so it's kind of the lowest common denominator within our family. We really haven't been able to open up much at all. We're pretty confident in the technology, so we actually got our kids in one of the trials. Medical clinical experimentation is tough to do, but they've given this thing to a billion people, my wife and I got it, and they've been doing mRNA in small scale for 20 years, so we're not worried something bad is going to happen. Our youngest, she's almost two, she got two shots of something. Hopefully it's not saline. We should find out the moment they apply for emergency youth authorization. And then because now they have a family relationship with us, they want to work with families because you have to drive in, fill out journals, and stuff like that. Some people aren't necessarily reliable in the trial, so they give us priority for our older two kids, four and six. Our oldest is getting her shot of something on Friday. But we still have our four-year-old, so we're still being very conservative. We went to the beach yesterday, but we were 20 yards away from the nearest person. We're still saying we're not playing with neighborhood kids, still staying local to the house and whatnot, so still pretty restrictive. And then we're just watching the cases and stuff like that. So that's kind of how my local situation is. Obviously, people have different situations, different risk tolerances, and stuff. But that's the sort of effects and choices we're making over here.

Yeah, so it's not me. I've got the philosophy degree, so even less connected to engineering. My wife's got the psychology degree; she's got three of them, she's a PhD. So in my office hours, we always talk about psychology, and I've been begging for the opportunity to kind of pull her into one of those because she studies psychopaths and stuff, and it's really interesting, hopefully not applicable at work, but a lot of people find that interesting. So yeah, I have a philosophy degree, liberal arts degree. I started college in electrical engineering, so I was going the technical route, and I just ended up being more abstract math on the chalkboard rather than physical robotics and stuff that I wanted to do. So I got frustrated, and I went to my advisor's office, and my needle kind of went to the other end of the spectrum: philosophy. But programming and computer science was another interest, and I almost ended up there. Luckily, I was programming as a hobbyist kind of along the whole way. So it certainly would have been, given what I do now, a shorter, less risky route to just get the comp sci degree and go into tech, as I ended up doing anyway. But philosophy taught me how to think, made me generally well read. I think if you look at engineering management at Silicon Valley-style venture-funded companies, you probably find an over-representation of liberal arts degrees like that: history, English, philosophy, a handful of others in engineering management, VP of Engineering, and CTOs and whatnot. I meet a surprisingly not a lot, but more than you'd certainly expect. So I think there's something to it, and I don't know exactly what that is. But obviously, any management responsibility deals with people. There's certainly a lot of communication that goes along with being an executive. So I think those fields of study tend to lend themselves to management in general. And luckily, I happened to get into tech, and I think it's still true today where you don't necessarily need a credential; you can just, if you can do the job, you can get the job, which is good. I think it's because most tech companies are fairly entrepreneurial. I hope it stays that way, but most fields stabilize over time, and eventually you need the specific degree to get the entry-level job and go up from there. But I don't see tech necessarily stabilizing or slowing down for a long time, and I think that's a good thing. Obviously, there's a lot of things we do at GitLab, like the ERG mentorship program and other things, where we're trying to bring relatively highly compensated tech jobs to places like sub-Saharan Africa and whatnot. I think that's a good thing. The tech industry can hopefully give people a rewarding career, in some cases lift them out of poverty, build a middle class in places that don't have it. So we're trying to do our part to follow through on some of those things. And me not going in the kind of straight and narrow route always provides a little bit extra push to do things like that because I was given some chances by some generous people, and I want to pay that forward. Pay it back? Pay it forward, I think, is the right direction vector. But thanks for asking.

Sure.

Anyone got a fifth question?

Yeah, I mean mostly what you'd expect: the technical, the engineering work. I'll say, having been in this role across several companies now, I always would just program personal projects on the side, and that kept me somewhat sharp, although not as sharp as people that do it 24/7 like Stan and Camille and Jagorshan and all them. But having kids, having young kids, really wiped the hobbyist stuff off my calendar for the time being. But as they get older, it'll be a little bit more possible to do things like read books and learn a new programming language or whatnot. So I know that stuff will come back. But it is a good thing to think about if you're going into engineering management. I encourage people to be deliberate about when they can take advantage of opportunities to check back in and refresh their technical skills. The last time I really did that in a meaningful way was two companies ago. I was a senior director, and the company was based on the East Coast, and I moved out to San Francisco. I was in a satellite office and took that opportunity to kind of downscale my responsibilities to a very small team. I took the opportunity to be a maintainer of a couple of apps and a couple of websites, and was on call and the main programmer. This was like 2011, 2012, when Node was new on the scene, so I learned server-side JavaScript. That was a lot of fun, and I still reference some things from those times when I get into technical conversations. The rest is just kind of like reading, which is not the same as doing. Reading about programming pales in comparison to actually doing it in terms of how it informs the decisions you make. So as I look ahead in my career, I always wonder when the next time will be that I get the chance to do that. Who knows what it might be, Rust or something that hasn't been invented yet, or if it'll just be the hobbyist stuff that comes back online when the kids get a little bit older and they're all in school and more independent.

Yeah, good question, and thanks. I do try to become cool and collected. I think there's really no substitute for just practice. So for something like this, the shameless plug for my office hours: I basically do an AMA once a week. It's open to the whole company, but it's on the calendars of everybody in engineering, and it happens either afternoon my time or morning my time so I can overlap with the APAC folks or the EMEA folks. It's a little bit more conversational than this, but it's pretty similar. So it's rare that I get a question I haven't literally fielded before in many cases. So I've kind of always thought through, if not a specific answer, a topic or something like that. I probably get challenged or caught off guard a little bit more in that setting because it happens so frequently. And then every once in a while, I do an AMA like this, and it's that much easier in comparison. Let's see, advice for people that want to talk to customers and stuff. I don't know, I just say look for opportunities. There's no reason, especially at our style of company, why you can't get started just shadowing meetings. It's interesting. I didn't do a shadow program, but I talked with Sid a lot, and Sid Scott, the shadows and people entering that program versus exiting it, just over the span of two weeks they come out saying, 'Gosh, I feel like I know so much about the company now, and I got exposed to so much stuff, and it was really intense but it was great.' And for the most part, there's very little requirement for them to talk and contribute, but just getting exposed to that stuff seems to increase their knowledge and confidence quite a bit. So that's definitely something to take advantage of. But at a smaller scale, there's no reason why you can't ride along with a TAM going to a customer meeting or someone like that and just start to see how people communicate, the cadence of those meetings. In my experience, when you do something like that, eventually there's a question asked where nobody knows the answer, and it's like, 'Oh, me the fly on the wall actually knows this one.' You take that opportunity to chime in, and that's your foothold to start contributing and participating. But also, at GitLab, one of our secrets is we keep the stakes pretty low for making mistakes. So I encourage you not to feel like you have to hold back or be perfect in those settings. That's a really slow way to learn in the scheme of things. So we try to have an environment where people are comfortable taking those small interpersonal risks. If you're wrong, no big deal, just find the right answer together and move forward and try again next time.

Sure.

Yeah, good question, Sean. I had a mental note to answer that in our question, and I forgot, so thanks for bringing that up. No, I was actually a really anxious little kid. I was the kid who had insomnia worrying about global warming and stuff like that. I was a wreck. I was a super straight-A student, everything had to be perfect, and I kind of burned myself out. I was probably not even a teenager and kind of burned myself out in school because I was holding on too tight. For me, learning to step back and realize things aren't a disaster was part of the step I addressed earlier about getting burned out on EE and going into philosophy. That was part of that journey for me. So yeah, I think for those of you that are concerned about this, at least one data point: yes, this can be learned. It's not innate. It's not like you're either one nature or the other. You can choose to adapt if you want.

Anyone got another one? Looks like we're up to number seven in the doc.

Cool. Well, we're coming up on time here, so why don't we end it here? But thanks everybody for the good questions and discussions. As I said, my office hours happen once a week, so feel free to pop in there. If you forgot something you wanted to ask, that's a good forum, and in some cases we can go deeper on the answers and have some back and forth on this stuff. Also, my staff meeting is on Wednesday, and that's open to the company as well. So if you're interested in listening in or contributing, just look at my calendar and add yourself, and we'll see there. So thank you.