Intro
Hi, this is CoRecursive and I’m Adam Gordon Bell. Each episode is the story of a piece of software being built. Today’s story is about Chef, the transformational tool that revolutionized infrastructure automation, the tool that was used everywhere from enterprises to tech giants, a tool that started as an open source project to solve specific problems, and then grew into this VC-backed company. That’s the story.
Well sort of, because actually the story is about a crisis. It’s not really about the tool, it’s about the people around the tool. And one person in particular, Adam Jacob, who founded Chef, and who ended up with a problem. Chef had hit a rough patch. Employees felt defeated. Morale was low. And so Adam had to rally the troops.
Adam Jacob: And so I wrote my little inspirational speech about why we weren’t dead and why what needed to happen instead was we just needed to work again. That all this ever was work, and you needed to ignore all the noise. And if we just kept doing the work with this amazing group of people and the way that we know how, we’re going to figure it out. And it doesn’t sound very inspirational now that I’m saying it. Just go with me for a second. Imagine it was inspirational.
And I got done giving this inspirational speech and I just barely made it off camera. Luckily I was alone. And just completely broke. I wept. The kind of weeping where you can’t stand up anymore. The office had beanbags in it because startups. And so I collapsed into a beanbag, and wept in a way that like top five all times so far life weepings.
Adam: Adam was crying, weeping, not because he didn’t believe in what he said, he did, but because the pressure was crushing him. When you go from a talented sys admin to the CTO of a big DevOps startup. You can get imposter syndrome real bad. But not just that. To understand that moment, that moment where Adam broke down, you need to learn about sys admins and the shit they have to deal with, and about the history of the commercial internet, how it grew, how it unfolded. How it went from dentist office ISPs to giant tech companies.
Because the way that Adam got to that moment where things seemed so bad, it started way back when he was a kid, when he was eight. When he was in the Pacific Northwest in his parents’ home, and he figured out that he could connect to bulletin board systems, to BBSs.
Bulletin Boards
Adam Jacob: I was a kid in a basement with bad allergies, and you could reach out into the world and meet strangers. And they didn’t know that I was a kid. I used to talk about politics and military strategy. I played lots of role playing games. Everybody who was playing would dial into the bulletin board, and the game master would write a short story. And you’d reply with a little short story about what your move was. The early internet, pre-proto internet was super fun.
Adam: That BBS game went on for years, but it eventually crossed over into the real world. They all decided to meet at the GM’s apartment for a real world game.
Adam Jacob: I get my mom to drive me over there, and I show up at the door and I’m like 12. And all these dudes are 20, 25, 30, I think the oldest was probably in his 40s. And I knock on the door. He opens the door up, I’m standing there with my little fucking bag of dice or whatever and some chips. And I’m like, “Hi.” And they’re like, “Who are you?” I’m like, “I’m Adam.” And they’re like, “We got to talk to your mom. Is your mom around? Because this got weird now.” Yeah, but they let me play because they’ve been playing with me for years. They just didn’t know.
Adam: Your mom okayed it?
Adam Jacob: Yeah, my mom totally okayed it. She was like, yes. I’ve been telling them who these people are forever. She knew all their names. She probably knew more about them than most of my friends. But that’s the kind of shenanigans that if you did that now. If my daughter came home and was like, “I made friends on Reddit,” you’d be like, “No.” But yeah, that happened all the time in the early internet for sure.
Adam: So Adam had fun. Dungeons and Dragons, making friends, playing online games. It sounds fantastic. But something else was a bigger draw for him. How did the bulletin board system work?
Adam Jacob: How do I make my own? How does FidoNet work? Before there was email, if you wanted to send a message to the East Coast, you could totally do it with FidoNet, but it happened because every night at 2:00 in the morning, your FidoNet hub would make a phone call to every single bulletin board in an area code. And then it would make one long distance call, and send all that mail to another long distance hub, and then transfer the mail. So it might take a couple of days to get all the way around the world.
And those systems are so interesting. You could dial into someone’s bulletin board, and then figure out how to do it. You could download bulletin board software. You could figure out how it worked. You could build the menus, you could install the door games. You could figure out how multiple lines work. So next thing you know, you’re like, oh, I got to get an operating system that allows me to multitask because I want to have more than one user at the same time. And then you’re like, oh, I got to get OS/2. So you go get OS/2 Warp and now you’re like running Warp, and that’s weird and nobody else is, and it costs a bunch of money. Next thing you run on Linux. And so you could figure it out. And it was so fun. And I just loved that exploration and I love that I could do whatever I wanted.
Dentist Office ISP
Adam: That era, the era of bulletin board systems, was slowly being replaced by the era of ISPs, internet service providers and the internet. And that provided an opportunity for high school age Adam. He started working for a local ISP based in the back of a dentist’s office.
Adam Jacob: You would walk past the dental hygienist or whatever into the room in the back that they converted from being an extra space for doing dentistry into a rack of modems and a Windows NT server and a BSDi box, and some phones for helping people get on the internet.
Adam: So would you phone the dentist, or they had their own lines?
Adam Jacob: No, we had our own phone lines. Yeah, you ran COTs lines. You ran phone lines into the back of a dentist’s office. So he had, what, probably a dozen phone lines run into the dentist’s office that then terminated right in a rack that he’d installed in the dentist’s office. And you plugged them into modems in the rack, and you sort of moved on. And the job interview was like, “Hey kid, do you know how modems work? Can you set up this rack of modems to connect people to the internet? And when they call you on the phone, can you help them figure out Trumpet Winsock or whatever so they can get logged in with Windows 3.11?” And I’m like, “Yeah. My favorite hobby as a kid was playing with operating systems and building bulletin boards.” And so I knew how to do all that stuff.
Adam: Then Adam met the sys admin.
Adam Jacob: He was like a goth kid that worked at the community college. He ran the community college computer lab, and the dentist had recruited him. And he always had better things to do than come to work at the dentist’s office because he was cool. Had a cute girlfriend and was trying to build a life or whatever. So for him, it was kind of a pain in his ass that he had to work there. And I was like, this is awesome. Yeah, sure, give me root. All I wanted in life was to be the systems administrator at that ISP because how fun, right? And yeah, that’s where my career started.
Mom and Pop ISPs
Adam: So meanwhile, while Adam’s finishing high school and being a sys admin, learning how to administer a BSDi box and thinking about whether college is an option, while all that’s happening, the internet is taking off. It’s going from just a few universities, few people getting email or using FidoNet to climbing numbers of real households, going from nobody to 10% of households in the US being on the internet. And all of these mom and pop ISPs are sprouting up. And Adam got a chance to work for quite a few of them.
Adam Jacob: I had a boss who was the pinnacle, I think, of the mom and pop ISP. The ISP was in his garage. And he was fascinating. He didn’t like to wear pants. He would wear pants, but he would remove the zippers from all of his jeans for some reason that I’ve never understood. And also the button, like the top button. And then he would just tie rope. He liked to have his pants tied with rope. He was also one of the people who originated the iridium satellites at Motorola, and then decided to go run this ISP.
His name was Jeff. He taught me to program. My job interview for Jeff was I came to his garage and the instructions were like, walk around to the side of the garage and knock on the door. And so I did. He opens the door. He’s got a phone tucked into his shoulder, like a landline phone. I have to clarify that because everybody would think cell phone, it’s not. Like an old red phone. And it’s tucked in the crook of his ear. And he’s like, “Yeah, yeah, come in.”
And he’s screaming at this customer. “No, shut up. Shut the fuck up. Shut up. No, it’s not my problem. The problem is you’re a fucking idiot. I tried to explain it to you. No, shut up. Fuck you.” Bang. Slams the phone down. I’m like, “Whoa, what was that?” He was like, “Hey, sorry about that. Are you Adam?” And I’m like, “Yeah, I’m Adam.” He’s like, “You can do tech support, ISPs?” I’m like, “Yeah, yes.” He’s like, “Yeah? You’re good at it?” I’m like, “I’m better than you.” And he’s like, “Okay, good deal. Get in the chair.”
Adam: After a couple years of this, Adam had a very valuable skillset. His job let him learn more about networking and operating systems and programming. He’s become sys admin/developer/customer support/whatever else. He tries college, but I mean, he’s a hot commodity doing this, so why bother. From here on though, the game changes. ISPs consolidate. Setting them up and running them, it’s been figured out. And so Adam shifts focus. He’s all about deployment now. Getting websites up, getting services up and running and on the internet. Because the internet, it’s got people on it now, and those people need things to see. Adam worked for Go2Net whose business model was mainly buying up various internet companies and squishing them together. His job was to get the acquired companies’ software up and running on their internal infrastructure.
Operations at Go2Net
Adam Jacob: So your job is to go figure out how it runs, port it to our infrastructure, and run it more efficiently. Especially so that then, we were public companies because anybody could be a public company apparently back then, so then you could show to The Street that you made more money because, even if you didn’t move the needle at all on the top line, the bottom line number could be significantly reduced because you’d gotten rid of a bunch of headcount. It ran on fewer servers. So we became really good at looking at random systems and going, great, how do I run this random system in production? How can I figure out how to automate it? And so we just got really into automation.
The French Canadains
Adam: This automation and consolidation work, it eventually brought him into contact with a voice recognition software team.
Adam Jacob: And the guys who made this thing were in Montreal. And they were very upset that they’d been acquired. They were not pleased that they’d been bought by InfoSpace, and they did not all leave because they were the only people in the world who had this particular expertise. And so I had installed this software they told me to install to manage the phone system. It was one of the first voice recognition phone systems. And it did not work well. You’d pick it up and you’d be like, “Call Adam Gordon Bell.”And it would be like, “Call Jane Smith.” And you’d be like, “Fuck, no, that doesn’t sound anything like Adam Gordon Bell.” And it was just every time.
I finally got sent to Montreal to integrate this company into our own, six or eight months after we bought them. Because they’ve stonewalled everybody else. And so they were like, “Kid, you got to go. Go do this thing and bring them into the family or whatever.” So I got on a plane, I go to Montreal, and I, through being a nice person or whatever, get them to like me enough that they decide to take me out drinking.
Adam: So the team takes Adam out drinking. Montreal, if you don’t know, is known for its active nightlife, it’s active bar scene. And after a couple drinks in, they say, “Well, should we tell him?”
Adam Jacob: And I’m like, “Now you have to tell me.” Because as soon as you say, should we tell them, you got to fucking tell me. They finally crack because they can’t keep it inside anymore, that the reason the phone system is so bad is that they had retrained the voice recognition software on their own voices. And they all had thick French Canadian accents. So they were, “Call the phone system and make a bad French accent.” And I was like, “What?” And they were like, “No, just do it.”
And so the CEO’s name was Naveen Jain. They were like, “Try to call Naveen.” And so I called the system, and I was like, “Call Naveen Jain.” And it didn’t work. And they’re like, “No, do it like you’re a French person.” And I was like, “Call Naveen Jain.” And it was like, “Naveen Jain.” It worked. And they just thought it was the funniest thing in the universe.
Adam: So the job was fun. Adam was good at it, but he didn’t always get the respect he deserved. For instance, this company gave him stock options, but there was a catch.
The Stock Options
Adam Jacob: Because I was a systems administrator, I was on basically a blackout list that was pre-done so that the executives could sell their shares. And in order to sell my shares, I needed permission from the corporate lawyer. So I went to the corporate lawyer who I’d known for a decade. Knew me, knew I didn’t know any insider data, and was like, “Hey, can you sign this piece of paper that lets me sell my shares so I can make some money?” And he was like, “No.” And I was like, “Why not?” And he was like, “Well, that would be a pain in my ass, and the other people who are executives in the company, it would be a pain in their ass too.”
And I was like, “Yeah, but you’re all real rich, and I’ve worked for you for a while. Wouldn’t it be great if, I don’t know, maybe you could just do this one thing that’s really not that big a pain in your ass because it’s really just you signing this piece of paper.” And he was like, “No, so you can leave now.” And I was like, “Okay.” And it really was as simple as signing a piece of paper, which was what made me angry. I wasn’t mad about the amount of the options. Do you know what I mean? I wasn’t mad about my salary. I didn’t feel like I deserved a bigger cut, but what bothered me was what felt like the injustice of it. It was like, look, I’m out here doing all this work for you because it’s good for me too. I love the people I worked with. What a fun job. The circumstances were interesting. There was a lot there that I really like. But again, it was this moment where it was clear that the people I was working for weren’t good people and that bothered me.
Adam: Also, it bothered him the way sys admins were treated.
System Adminstrators Day
Adam Jacob: So systems administrators used to have, maybe we still have it, but we don’t even have that title anymore. We’re all DevOps engineers now, or cloud engineers or platform engineers, or whatever we are. We’ve jettisoned the title. But the reason we jettisoned that title was because we were so loathed. Systems administrators had an appreciation day. And a thing that’s real about people who need appreciation days, it means that every other day, they’re not appreciated. That’s just sort of how it works. If you’re appreciated every day, then you don’t really get an appreciation day.
We had a boss, a guy we worked for, he’s like a skip level boss, who took us out for beer. And there were maybe 20 systems administrators at the time on the team. And took us out for beer on systems administrator appreciation day. Brings the beer over, sits it down, he’s like, “Happy systems administrator appreciation day. Too bad none of you’re smart enough to be software developers.” That was the big thank you moment.
Adam: So Adam’s underappreciated, right? And this stock option thing and other factors, they make him want to leave Go2Net. So him and his friends start a sys admin consulting company. And as they’re handling various clients, Adam creates a tool to make this type of work easier. This tool is Chef, and fast forward through a lot, and Chef becomes not just an open source tool that Adam built, but a whole dev tool startup. The cool thing is Adam gets to be CTO now. He gets to build a team around the product that he built for his own use. But even cooler than that, he gets to be the champion for sys admins everywhere.
SysAdmins Are Programmers
Adam Jacob: In the early days of Chef, part of what Chef did was say that people who were systems administrators were programmers. And that actually all of them were programmers and always had been. And that the distinction between software developers and systems administrators is actually non-existent. It’s just specialization. The fact that I specialized in systems, and maybe you specialized in application development, we were both equally skilled and we were working in the same problems. We just had different specializations. And that the people that did that work were capable programmers, and in fact should write programs, they should write code to define how the infrastructure should be automated.
Adam: So this becomes the infrastructure as code movement. Adam is winning hearts and minds. He’s going into large enterprises telling them, showing them how to move into this new world, and they’re doing it. And they’re doing it using his tool, and they’re doing it buying licenses for Chef. And so Chef is making money, and it’s kicking ass.
Adam Jacob: And I had a meeting with, that’s how I met one of my good friends, Jamie Winsor, but we had a meeting with this company that Jamie worked at. Jamie had invited me in because they started using Chef and wanted them to buy a license. So I sat down at the table. And I was on one side of the table, and the other side was their boss and six systems administrators sort of lined up beneath him.
The boss’ opening question for me was, “I don’t understand why we’re even talking about this because my guys are too stupid to learn to program.” And said it in front of them. They’re lined up like little ducklings. And there’s your boss being like, I got the dummy crew here, man. Why are we even talking about this? You could watch everybody deflate because what a deflating thing to have someone say about you.
Adam: Maybe because of his history or just who he is, Adam’s really sensitive to this type of stuff. People trying to make themselves feel better by pushing others down, and so he handles it well.
Adam Jacob: I just flipped into self-righteous fury. I was like, “Look, for sure that’s what you’re going to get if you treat them that way. If you’ve talked to me like that, I’m not doing shit for you. These are smart people, fully capable of doing this work. They’ll kill it for you, but never if you act like that. If you act like this, if this is who you are, A, I don’t want to sell it to you, so that’s easy. But B, you’re never going to win treating people that way. You just won’t. You can’t. Of course they’re not going to do their best work for you, just beating them down, grinding them down all day.”
Adam: Adam was a natural defender of people, especially operations people, and that helped build a following behind Chef. But as the company succeeded and as it grew, being a sys admin defender wasn’t the job anymore. Even perfecting Chef, the tool he originally built, that wasn’t the job anymore either. As a company grows, the things you need to do in it, they change.
Struggling With Being a CTO
Adam Jacob: You have a responsibility to the people that you employ and to the community that you build around the software that you don’t have when you’re just an employee. And it’s heavy. Last week you were a systems administrator, and you knew you could run a little consulting company, but you didn’t know you could run a startup with $2.5 million. I didn’t know that I could walk into Facebook and talk to their infrastructure automation team, and be as good or better than they were at what they did and help them understand how to design automation for Facebook. You don’t know that until it happens. You want it to work, not only for yourself, but for everyone that works for you.
And you are responsible for them. You’re responsible for their families. Their kids are going to go to college hopefully because whatever you’re doing works. Or they’re going to buy a house. That’s their life. And a lot of the messy parts there are also that your own identity as an executive or as an entrepreneur, it doesn’t exist really at first. You don’t know that those things are things you can do. And the thing about it is that it causes a lot of stress on your own identity. It causes a lot of sheer in who you are as a person.
How Not To Be In Charge
Adam: This is imposter syndrome, right? Adam felt like he was pretending to be a CTO, and that he’d be found out. And if you think it’s tough as a developer to have imposter syndrome, imagine the stress when you’re suddenly a CTO of a growing company. Here’s what’s strange though, the best leaders, in my experience, they always seem to come from the ranks. They did the work, and now they lead. But it’s not clear to me how that happens. There’s not that many skills that overlap between being an engineer and being an executive, being a leader. You start as sys admin, and then you read some book of business platitudes, and then all of a sudden you’re a visionary executive quoting Steve Jobs. I don’t think so, right? I don’t think that’s how that works. I think you have to figure out how to lead and what type of leader you are. You have to figure out how to be in charge. And that is Adam’s ultimate challenge at Chef.
Adam Jacob: I think we tend to believe that when we become in charge, that when we become the executive, or we become the boss, or we become the product leader, or the engineering leader, whatever it is, that suddenly what’s going to happen is it’s going to work your way. That now finally things will go my way. And I think the truth is that they can. One way of leading people is in fact to just make them do what you say. And the more tyrannical you are about it, there’s some interesting workplace studies that kind of back this up, that being a complete tyrant is a very effective way to get a lot of work done. And also it’s not very good for the people that you’re tyranting around and they tend to not last very long. Long-term, it’s not particularly great, but short-term, whip the horses. They’ll run.
Adam: Adam wasn’t the crack the whip type, but he was still figuring things out, and the stakes were high. They weren’t the first infrastructure as code tool. Puppet was pretty popular and things predated Puppet, and then cloud was happening and newer tools were popping up and competition was everywhere. They had to get things right, and Adam was still figuring things out, and this led to a lot of arguments.
Adam Jacob: You really believe that there’s only one right way. A good example is what’s the right shape of an API endpoint look like? Or how’s this feature supposed to be implemented? The conversations I would have would get really tactical really quickly. You’d wind up in conversations that are like, “No, obviously the right API shape is this one.” And you’d be like, “Why?” In the end it would be because I said so. You boil all the arguments down, and the answer is because I’m your boss and force.
Adam: Would it come to that, or would it just be kind of implied because you’re in charge?
Adam Jacob: No. It would come to that because I was also trying to be a good person, which I know sounds silly. But the worst person to work for is the waffley one where they’re 100% going to use hard power to make you do what they want, and the whole time they’re pretending they’re not. They’re pretending the whole time that you have a say in the matter. They want to know what you think, that they’re listening to you. And that if your argument was good enough, you could convince them of the truth.
But in reality, they always knew what they wanted. And they also, because someone told them they should listen, they’re listening to you because they don’t want to be a piece of shit. So they’re like listening, but they’re not actually going to change their mind. They’re just waiting for their opportunity to tell you what to do to get you back to work because you’re a pain in their ass because they already know what they want. And I don’t love that that’s who I was. Do you know what I mean? That’s not my favorite version of myself, but it’s what I was doing.
Adam: Adam says the thing about conversations like that that he knows now, but he didn’t know back then when he was in the heat of it, those conversations where you’re arguing about the small stuff, arguing about REST or about RPC, they’re never really about the details. They’re about something else.
Adam Jacob: The odds that today is the day that you’re making the singular make or break decision for the next 10 years of your work, pretty fucking unlikely. But boy, you’d argue about it like you were. You’d fight about it. It really was every argument was about gnawing your arm off about whether or not you were who you said you were. Everything is vitally important all the time. And if any of it’s wrong or fails, then the whole thing falls apart, and now suddenly you’re a fraud. You’ve let everyone you know and love down. That’s a lot to carry around with you to work every day. If that API shape wasn’t right, the whole fucking thing falls apart.
Because again, riding on it was not just that API call, it was my identity. If it failed and fell apart, what did that say about me as a person? Does that mean I’m a bad person? Does it mean I never belonged here? Does it mean that everything I’ve been saying about who I am and what I want and what I can do, were those all lies? It was lot of pressure on what is essentially should it be RESTful or RPC over JSON? And who fucking cares? Who fucking cares?
Adam: There was a lot on the line here, jobs, for instance, people’s livelihoods. But also if I’m understanding Adam right, his need for job confirmed self-worth, it made it seem like everything was important. This is maybe a different side of feeling like you’re not cutting it. If you don’t trust your abilities, if you’re worried you’re an imposter, it can actually kind of make you a difficult person to work for.
Adam Jacob: Oh my God, I hope people have liked working with me. I feel like they have, but there’s definitely people who maybe wouldn’t do it again. Or if you ask them what it was like working with me in those moments, in the era that I’m describing, they’d tell you how hard it was, and how hard I was, and how it wasn’t the best.
The Motivational Speech
Adam: Which brings us to the motivational speech. Chef as a tool, as a company, it was a success, I think. Just imagine how hard it is to build a tool and have it become this successful, to become so popular that a community forms around it, a company forms around it. Hyperscale and companies like Facebook are using it to manage their giant data centers. But usage isn’t really enough. The game is to be the dominant tool, to build a natural monopoly and build your company around that. That looked like it was working at first, but things change.
Adam Jacob: Ansible showed up and then Ansible was cooler than we were. And then Docker showed up, and it was cooler than everybody was. And then Kubernetes showed up, and it was cooler than Docker. Meanwhile, cloud’s happening.
Adam: Chef was used in lots of places. It was a great and practical tool. Lots of enterprises were probably improving how they worked because of Chef. But if the buzz in the industry changed to Kubernetes or whatever, and it’s a startup where everyone’s looking for meteoric growth, up and to the right, then things can start to seem bad.
Adam Jacob: I had wound up in a meeting where I needed to give an inspirational speech. Like people were feeling down about what was happening and about our prospects and about whether we could succeed. I had been doing research to figure out what do good inspirational speeches look like because I want to be good at it. So I was rewatching, and have you ever seen Any Given Sunday? It’s an Al Pacino movie where he’s a football coach, and he gives an incredible inspirational speech at the end.
Adam: So Adam gives a different but similar inspirational speech because it’s part of his role. If the investors, if the community, but especially if the employees, if they don’t connect to the vision, then you’re dead in the water. So he gives his speech, he motivates them, tells them that the key to success is just doing the work.
Adam Jacob: And if we just kept doing the work with this amazing group of people in the way that we know how, we’re going to figure it out. But we just needed to come together and do more work. I think in startups in general, we tend to mythologize everything but the work. We mythologize the founding stories, we mythologize the outcomes. Mitchell Hashimoto and Armon Dadgar made HashiCorp, and now they’re worth a ton of money. And Mitchell writes his own terminal emulator in his spare time, and learned to fly a plane. And Mitchell is a delightful human being. I like him a lot. I’m so glad for his success. And also it’s really easy to mythologize Mitchell.
And the most impressive thing about Mitchell is none of those things. The most impressive thing about Mitchell to me is that Mitchell stayed in that company, still is in that company, and found a way to be useful and to grow and to change as that company needed him to grow and change. So did Armon. And they learned how to do the work, and to keep doing it, and to be good at it, and to stay good at it over a very long time through a huge amount of change.
But we don’t mythologize that. We mythologize that Mitchell wrote Vagrant. And then from Vagrant in his dorm room, now he’s married to a movie star, airplane terminal emulator, and then in the middle was all this work, and we just skip over it. We just fast forward through it because it’s not a very good story because it’s like, well, what’d you do? Well, I worked a lot.
Adam: Yeah, this is the montage part with the music.
Adam Jacob: I love the non-montage part. I’m a little obsessive about how much I like the work of it. I just I like the work. All the rest of it is the explanation. But it was the work that did it, right? It was all those people who showed up every day and decided to believe in whatever it was you’d ask them to believe in. The work is good and valuable in its own right, and actually the only thing that ever mattered.
Adam: The speech went well, right? Adam is a great speaker. And so the team is excited to do the work of continuing to build Chef. But as soon as the speech ended and Adam gets off camera, things change.
The Break Down
Adam Jacob: I wept. The kind of weeping where you can’t stand up anymore. So I collapsed into a beanbag and wept in a way that like a top five all times so far life weepings. Not as badly as when my father died, but in the room of weeping, real emotional stuff. The speech wasn’t a lie. That wasn’t the issue. I was so scared that it wouldn’t work. And that I was going to let all those people down, and that I was going to let my investors down. But mostly I was worried about the people that I worked with. I was worried about the people who worked for me. I was worried about all those people who had dedicated their lives to this thing that we decided to do. And I felt so afraid and so false.
And I was like, what if someone saw me weeping in this moment? How do I explain? How does it not look like I’m weeping because I was lying? And I wasn’t lying. And I realized in that moment that it couldn’t be about my identity. That whether or not those people that I needed to take care of, I do have a responsibility to those people. My responsibility to those people is not that we don’t fail. My responsibility to those people is that that I’m going to outwork them. We’re not going to fail because I’m not going to do the work. If we fail, it’s because it just wasn’t going to work.
Adam: So Adam basically took his own advice about doing the work. But it’s more than that. This was the moment where Adam figured out how to be a professional. He escaped this identity crisis. He escaped this imposter syndrome. He was basically forced to. He was forced to separate his work from who he was because he had no other choice. The alternative was being crushed by the pressure of the startup game.
Adam Jacob: I couldn’t go on with my identity being tied to the outcomes of this capricious monster. Because it is a monster. It’ll take whatever you put into it, and never give you anything back. It never tells you that it loves you. It’s not going to come to your birthday party. It’s not going to hold your hand on your deathbed. It’s not. So you have to be the one that figures out how to put limits on it, and you have to be the one to figure out how to professionalize it so that it can be work. Because if I wanted to serve those people, and continue to serve them and actually hold up my end of those obligations, I couldn’t stay insecure like that anymore. I didn’t have the luxury in some ways of making it all about me, which I really, really, really wanted to do.
Adam: Wow, that’s powerful. It’s like you’re not putting it in a box.
Adam Jacob: No, you’re just accepting it, and accepting that it’s out of your control.
It’s Out of Control
Adam: The realization that this was out of Adam’s control, it changed him. Suddenly everywhere he saw people wrestling with control and with identity and with who they were. But now Adam’s given up on control. He doesn’t need to prove he’s CTO material by diving deep into every technical discussion. He just sets the direction. He does the work, he trusts his people. It’s similar to the transition we all go through when we do our first real job.
Adam Jacob: Before you’d had your first job as a software developer, did you believe you could be a professional software developer?
Adam: Yeah, that’s a good question. I guess not.
Adam Jacob: I mean, maybe you knew a language or whatever, but you didn’t know you could be a professional. You didn’t know you could get paid for it. There’s a whole lot you didn’t know about yourself. And over the years of doing that profession, you became a professional software developer. Where now if somebody asks you, “Are you a professional software developer,” you’re like, “Yeah. I got all this technique. I’ve seen all these things. I’m seasoned. I’m a veteran software developer. I know my craft.”
Adam: In other words, the work is the secret to the identity problem. The work is all we have sometimes. You can’t treat every bug like it’s a test of your ability. Every task can’t question your self-worth or you’ll crumble. But you also just can’t stare in the mirror and tell yourself you’re a professional and be more confident. Most people are smart enough not to buy their own bullshit. Instead, you have to generate evidence for yourself. You have to do the work. And that’s what happened.
Chef continued on, it got acquired. And for the VCs involved, Adam says it wasn’t a grand slam, but it was a double. It was a pretty good result. And that was 15 years of Adam’s life, right? But he’s not done. He still sees a lot of ways that things need to be made better.
Making Ops Better
Adam Jacob: All the decisions we had made since I started working in ISPs about how to put together infrastructure and internet applications and deliver them over time, and do CI and CD, and pipelines and infrastructure as code, and continuous delivery and all that stuff, when you put it all together, what we get are mediocre outcomes in general. We get month long deploy cycles. There are exceptions, but they’re the exception. And if we wanted it to be orders of magnitude better, we needed to redesign the system from scratch, that some of the variables needed to move.
And so when I left Chef, that’s what I knew I wanted to do. I was like, I have enough money and enough security and enough of a reputation that I can sit for a minute, and with my co-founders, think about if it was better, what would it be like, and how would it feel?
Adam: And now, because of all this hard work in his new company, in System Initiative, Adam is not a sys admin pretending to be something else.
Being Who You Are
Adam Jacob: I’m not pretending to be a CEO. I’m not pretending to be a person who can raise money. It’s not a stretch. I’m not hoping that I can figure it out. I know I can figure it out. I know I have those skills. I know that those things are there because I have done them and I am a professional who does that. And I haven’t always known those things. I had to learn them by deciding that was my job. And for a lot of that journey, I didn’t know it was my job. I thought it was my calling. I thought it was my responsibility. I thought it was a lot of things, but it wasn’t my job. And if I failed at it, I felt like I, Adam Jacob, was going to be a failure personally.
Adam: There’s an XKCD comic where he’s at the bank and the guy’s like, “Okay, any questions before we sign the mortgage?” And he’s like, “Well yeah, my question is I feel like I’m not really an adult even though I’m an adult age, and there’s no way possible that you should ever give me a mortgage. I’m really just a child inside. And nobody’s realized that before.” And then the guy’s like, “Well, any questions about the mortgage specifically?”
Adam Jacob: Exactly.
Adam: I feel like a CEO and entrepreneur can never say that out loud.
Adam Jacob: But I think it’d be better if they did. I don’t control whether System Initiative is going to be successful. I can only build the best art I can build. I can make the best thing in the universe that I can imagine building to solve this problem. Look, I’ll be judged in the end. And if System Initiative works out, then I’m a genius and the whole thing was brilliant. And if it doesn’t, then I’m a dummy, and there’ll be 100 reasons why I’m a dummy and it’s my fault. Then it’s fine. But in the middle, I’m doing really good work. If you looked at the work I’m doing, I’m doing good work. I’m doing a good job at the work. And yeah, I think the work is what matters.
Adam: Here’s the thing, I didn’t expect Zen monk type answers before I started chatting with Adam, but man, I feel like he’s pointed out some insecurities that I have, that we all have, and he’s given a solution. I think what he’s saying is that you can’t really control your life, and you just have to accept that.
Life Is Uncertain
Adam Jacob: People want to believe that it’s science, that if you do all the things the right way, that the results will come. And it’s just not true. There’s so many things you can’t control. There’s so many things that are subjective, that are about how people look, think, or feel about things. And that ultimately, the sooner you get to a spot where you realize that that lack of control over the outcome is a gift that allows you to just focus in on the work and just make the best art you can make, and not be held back by all the other concerns that you may have about its potential or its future.
And that that’s actually the best way to build good art. And it turns out that that’s the best way to build good music. It’s the best way to build good anything. It’s also the best way to build good companies. It’s also the best way to build good software. The sooner we figure it out, the happier we get to be because so much of the agita is wrapped up in the question of like, well, but am I really? As soon as you make art and put it in the world, you don’t have to ask the question anymore about whether you’re really an artist. Am I really a musician? Am I really a software developer? Am I really a systems administrator? Am I really a CEO? It’s the work that makes you those things.
We spend a lot of time in our world and in our lives feeling really stressed out about whether or not we’re going to be the people we say we are or the people we hope or want to be. There is a way, there’s a path to figure out how to just be the person that you are, and to be happy with being that person. And to then let the work be what pulls you into all the other things you want to be. And to free yourself a little from having to worry so much about whether or not you’re going to be successful or not successful.
Not About Winning
Adam: It’s been a bit since I talked to Adam. I’ve been thinking a lot about his advice. Do the work, be who you are. When I condense it down like that, it sounds just like a platitude, or the world’s least surprising headline. Startup CEO really likes work. But today I was reviewing a design document. I was leaving a comment, and I noticed that I wanted to win. I wanted to be right. I wanted the decision to go my way. That would make me feel good. And there was a flip side of it, right? A fear that I would be wrong. There’s an obvious problem with what I said, and it would go the opposite way. And then I thought about Adam and I thought, this is my identity. I want to be reassured that I’m good at this, right? But I don’t have to. It doesn’t have to be about lifting up or tearing down who I am. I can just do the work. I don’t have to win.
Outro
So thank you so much. Adam Jacob, sys admin, hard worker, CEO. Find him on Twitter @adamhjk. We didn’t actually get to talk that much about his new venture System Initiative, but you should check it out. I’ll a link to a fun talk where he goes through some of it and shares some fun stories because he’s just fantastic at that. If you like the podcast, check out my newsletter, comes out around the same time. It’s usually related to the episode. I cover similar topics. Or follow me on Twitter @adamgordonbell, where I share some of the behind the scenes details of the podcast as I’m working on it. And for true fans, here’s how you can help me. Go to corecursive.com/supporters, there’s a link in the show notes, and become a podcast backer. You’ll get bonus episodes and you’ll be part of the community.
And until next time, thank you so much for listening.
Sticher
Overcast
Apr 02, 2024
57
min