30: Ep. 30 | What all developers should know with Thomas Dohmke

Brittany Ellich (00:00) Welcome to the Overcommitted Podcast, your weekly dose of real engineering conversations. I’m your host, Brittany, and today I’m joined by the rest of the crew.

Bethany (00:09) Hey, I’m Bethany.

Erika (00:11) And I’m Erica.

Brittany Ellich (00:13) The three of us met while working on a team at GitHub and quickly realized we were all obsessed with getting better at what we do. So we decided to start this podcast to share what we’ve learned. We’ll be talking about everything from leveling up your technical skills to navigating your professional development, all with the goal of creating a community where engineers can learn and connect. Today on Overcommitted, we are joined by Thomas Domke, the former CEO of GitHub. And he recently left to go back to founder life and is currently working on a stealth startup.

Thomas Dohmke (00:42) You

Brittany Ellich (00:43) Thomas hasn’t just led one of the most important platforms in the world, not that we are, you know, biased there. He’s also a developer at heart with a passion for helping engineers grow, ship better code and stay curious. We’re diving into his journey, the lessons he wishes every developer should know and where he thinks we should all be focusing our energy in this fast moving industry. So let’s get into it. Welcome Thomas.

Thomas Dohmke (01:06) Hey everybody, and I loved how you made air quotes for those without saying air quotes to stealth startup. It’s actually a stealth startup. So great to know, but like, you know, it’s a proper term in the sense of, you know, not ready to talk about what the startup is doing, but anyway, so great to be here and to be with Hubbers again. It’s been two months, a little bit more than two months that I announced my departure. And so it’s great to be with you all today.

Brittany Ellich (01:10) You

Okay, is that the title?

That’s awesome. Yeah, thank you for joining. Yes, I did not know that that was an actual term. We’ve had a few startup founders recently on the podcast because none of us obviously work in the startup world currently, and it’s been fascinating to learn a lot about it. It’s been really cool. So what was your overall journey in working into software? What’s your story up until now?

Thomas Dohmke (01:49) It’s a long story, so let me make it quick so nobody gets bored about hearing this for too long. I grew up in East Germany when Germany was still divided into two countries. In East Berlin, in the 80s, I couldn’t just go and buy a computer because they weren’t available in stores in Germany. But I found computers with a friend of mine in the geography lab of all places in school. was a…

East German clone of a Z80 computer, so like a microcomputer. And we started coding in basic. I think it was still had a cassette tape, if I remember correctly. And then the wall fell and I got a Commodore 64. And as most kids, you know, that get into coding, I started just playing around on a mix of playing computer games and then when mom and dad said game time is over, then I was like, well, I’m coding. So that’s not really screen time because I’m learning something. And, know, back then the Commodore 64 was connected.

to little tiny TV, right? Like we didn’t have these fancy screens that we have right now with high resolution. It was like a crappy TV that you were looking into. So I could actually make an argument that you are way too close in front of that small, you know, whatever 10 inch TV or so. you know, learned, taught myself basic and bought books and magazines, got a PC, taught, learned more, and then went to university and learned a lot about, you know,

the internet at that time, because this was the new thing. And I had, think I had an AOL account first and school had some access, but it was super slow and super crappy. But it also was like this wide open world where all of a sudden you could chat with other developers and learn about how they’re doing things. And the first open source I had was Linux, SUSE Linux, I think 5.3 or something that I bought in a bookstore in Berlin. And so, you know, from that journey, we, I,

I always wanted to become a software developer. I knew this was like my passion. it was so cool to create something out of thin air, right? Like you didn’t have to buy a 3D printer, which wasn’t a thing either, but again, it didn’t have to buy a machine. You didn’t have to buy, you know, complicated equipment. You didn’t have to hire people. You could just start coding, build something, share with other people. And that passion kind of like has followed me through my career, whether it was in the automotive industry at Mercedes and later at Bosch or whether it was my own startup called HockeyApp that Microsoft acquired.

And then for the last seven, seven or so years at GitHub, um, where I think, you know, I had found my dream job. It was being the CEO of a company, uh, that develops software for software developer, kind of like the Home Depot in the software development world, right? Because I think most folks, when they go to Home Depot, they, they appreciate not only that they can buy tools, they also find other nerds. And that’s, that’s in a way what GitHub is today.

Brittany Ellich (04:19) I love that description, that’s amazing. I’m gonna steal that to describe it to other folks. I feel like everybody that I talk to that I work with, or that I, you in my family, they either know, like they ask where I work and they either think it’s like, like GrubHub, or they’re a software engineer and they’re like, yeah, I know GitHub, everybody knows GitHub, so it’s big divide I like that.

Were there any pivotal moments in your growth to where you were like, hey, this is the career for me. This is the thing, like a project or a mentor or anything that, you know, stands out to you.

Thomas Dohmke (04:49) A bunch of those, actually you mentioned mentor, that was probably the first one. So I was in university in Berlin building software for robots that would drive through the building. And that was, you know, the late 90s, early 2000s. So the robot was like, not that fancy at all. looked like a half a table, with sensors stacked on top and computers, you know, in the rack on the bottom, like literally like, you know desktop computers as part of the robot.

And I had a mentor at back then, Daimler-Chrysler today, Mercedes-Benz, and he visited me in university and was like, why don’t you come to Stuttgart, which is, you know, about 700 kilometers away from Berlin and southwest of Germany. And instead of building robots, you can write software for cars for the Mercedes S-Class at the time and have it drive around autonomously. And I was like, yeah, that sounds cool. And so I moved my life from Berlin to Stuttgart and joined Mercedes at that time. And I knew nothing about the automotive industry.

other than what I’ve heard about my mentor and obviously had been to like a factory seeing how that works. And that’s kind of like a to-do thing if you are the bucket list thing, if you, if you have a mentor and you get kind of supported by an automotive company. And so I think that was a pivotal moment because all of a sudden I came out of my university environment into, into, you know, real life and software development and cars, especially back then, or in the early 2000s was very basic. Like the, the

control units had a CPU and memory as little as the Commodore 64, even though it was now 10, 15 years later that most people would use a Commodore 64. Every button, every user interface element in cars back then was a question of how much money does it cost to add the button and how much complexity does it add to the car? Today it’s much easier as most cars have a central screen that is like a computer. And so I learned a lot about that. I also learned that

That’s not for me for in the long run. And so in 2008, when Steve Jobs showed the iPhone SDK, you know, about a year after the iPhone had come out and I was an original iPhone user. That was, think, the second pivotal moment of my career because I decided I had enough of the automotive industry. was done with my PhD thesis that I wrote about automotive control systems. And I was ready to move on and that there was the financial crisis, you know, in late 2008 and that didn’t stop me.

from deciding I want to become an independent software developer. And so I left my well-paid job at Bosch to jump into the life of an indie developer building apps for mostly German companies. In that day and age, most of the apps that you would build as a consultant was like tiny apps, finding gas stations or like a front end to online banking. But it’s really just a frame to web-based experience and what have you.

But that, you know, in itself then gave me the opportunity to build HockeyApp, which was a mobile development platform which got acquired by Microsoft. You know, that in itself was a pivotal moment as it moved my family and I from Stuttgart to Seattle. And then obviously the GitHub deal in 2018 when not only, you know, did we decide at Microsoft to buy GitHub, but my former manager, Nat Friedman, who then became the CEO, pulled me into that deal.

And kind of convinced me that my career at Microsoft as a director of product with, I don’t know, 25 or so people working for me was not as important as, you know, joining him with the GitHub leadership team through the deal and just, you know, jump into the cold water of that adventure of an independent GitHub within Microsoft.

Bethany (08:13) That is really awesome. It’s so cool to hear about the journey through different industries and how that all leads up and builds upon your experience and how you act in a certain role. I’m curious if there are any skills or habits that, through these industries, might differentiate the truly great engineers or the folks that you really enjoyed working with.

Thomas Dohmke (08:37) The first thing that always comes to my mind, if I look back on the 30 years or so that I’ve been coding and fun from one another, it’s the passion for doing that and passion for software development. It’s not always easy. One of the jokes I’ve made a lot over the last three years is that the happiest day of a developer is when they start a new project, file, project new, and then it’s all downhill from there. And in that moment, you’re excited. Everything is greenfield. You can try out a new language or a new component, new open source library or new design pattern.

And then, mean, you all know what that’s like four weeks later, you’re like, shit, you know, this doesn’t actually work. And the technology is not as mature as we thought it would be. And then you have to deliver the bad message to your manager that you effectively wasted four weeks or wasted, you know, air quotes. Obviously it’s obviously a learning opportunity. But I think the passion is what defines this role as the industry is moving so fast all the time. Right. And I think that’s number two, you have to constantly learn.

There is no stable state unless you want to work on COBOL and mainframes for your whole career, which you certainly can. I think most developers want to stay at the top of the technology. That means learning, learning, learning. And I remember leaving university and thinking, I’m done. I can focus on doing stuff and building and not so much on learning. think that’s like, you already quickly realized that’s not actually how the industry works. And soon enough you are surrounded by people that are

hopefully smarter than you and tell you, you know, how to do certain things and teach you things. And you’re like realizing, okay, I need to start reading certain blogs or, you know, catch up on the technology or maybe get over myself and figure out that the programming language I love isn’t the one that the team is using. And I don’t need to argue about that. Maybe you argue with somebody else on this podcast about who we were the school or something like that. But I think learning is super important. And then

You know, for me personally, there was always a bit of what decisions do I make that may be counterintuitive. I mentioned already one, which is like leave my career as director of product management at Microsoft behind to become VP of special projects at GitHub, which really was a role that didn’t mean anything at the time. It was just a question of, I join with Nat, the leadership team? And he was like, we will come up, you know, with the things that are important for the company to push forward. And it’s going to be fun. trust me, know,

He said to me, trust me, it’s going to be fine. And obviously, three years later, when I became the CEO, he proved himself right, or he proved it to me that he was right and that it led me on the right career path. But I think it’s these making these decisions at a time when you don’t have all the information, when you have to, to a certain degree, trust your gut. And that’s true for a career as it is true for architecture. You can have all the arguments in the world of whether this is going to scale to a hundred million users or a billion users and so on.

It doesn’t really matter. You’re always going to run into scale challenges. You’re always going to run into outages. I mean, we all know what that is like on the GitHub site. so you have to trust your gut and you have to trust your feelings. And then of course, you have to build out your experience and craft.

Bethany (11:31) That definitely makes sense. It definitely seems that that passion and the learning is just so important for continuing and staying relevant in the industry. But also just when you enjoy your job, it is so much easier to do well in that space. So that definitely is good advice.

Thomas Dohmke (11:44) Yeah.

Or, you know, just wake you up on a Monday morning and be motivated to go into your job when you have to pick up, you know, an item of from GitHub issues that isn’t as exciting as, building a new feature. mean, you all have been in different teams at GitHub. It’s not always the most exciting stuff that comes along in any given week. And so if you’re passionate about software development, if you actually enjoy, you know, writing software, it doesn’t really matter what feature you’re building, right? It matters that you enjoy doing it.

Bethany (12:14) Absolutely. It definitely does feel that way to me and I don’t know about the others, but I think it was just so cool how people use GitHub. It means something to people even when they’re upset through issues and things like that or there’s something wrong. It means people truly care about the platform that we’re building and believe in it. it is a cool thing even when there are those rough issues and rough days, for sure.

Thomas Dohmke (12:26) Yeah. Yeah.

Bethany (12:40) I’m curious if, I mean, being in so many different spaces of what you’re working on, do you have any advice for people who might feel stagnant in their career or who are stuck at a certain level or in a certain niche and might want to break out of that?

Thomas Dohmke (12:55) Yeah, I mean, I think, you know, it’s, it’s always challenging if you feel like you’re stuck because you have to convince yourself that there’s something else there first, right? That’s the first rule of having a growth mindset is to admit to yourself that we, every morning or so that we have some degree of static mindset. We are all, you know, being raised in some form of you’re good at something and not good at other things, right? Like you hear that so much and you hear people talking about themselves in that same way. And that’s a clear sign of you having a static mindset instead of saying,

Okay. I can improve myself. I can, you know, get out of this situation where I’m stuck and I can get over myself and maybe have a conversation in a secret Slack channel, or I have a conversation with your manager and, and, you know, be open-minded and saying, Hey, I feel like I can no longer progress here. what advice can you give to me? And I hope, you know, for most folks, that is the first step that they’re taking.

without looking for a new job in a different company or in a different team to have that open conversation with the manager. So that relationship that is so crucial is given the right amount of trust. And the manager might actually be cool and say, yeah, maybe a billing team has new opportunities for you or the Azure AI Foundry team over at Microsoft is the right place for you. And look, I’ve been through many stations in my own career from

a student to a big automotive company, to a big automotive supplier, to a startup, Microsoft and so on. It is cool, know, it is absolutely valuable to switch jobs and see what life is like. But you also need to realize that every next company they’re joining has their own set of problems. And they might not show those problems to you while you’re interviewing, but certainly you will find out in your first or second week, because it’s the nature of our jobs, whether that’s in engineering or leadership is…

Your job actually is to solve problems. Building a feature is just a different angle on a problem. You know, it’s a nice problem to have kind of like, but the most of the time we are solving really hard problems and you’re never going to run out of problems. You’re only going to hopefully at some point retire. But everybody deals with their own set of problems. That is the nature of, you know, the work that we do.

Erika (14:58) Let’s take that idea and turn it into kind of a hypothetical of if you could go back and give your former self like any advice about growth, development, learning software, what comes to mind for you for anything that would have been helpful for you to hear from what you know now.

Thomas Dohmke (15:18) think most people that go, you know, through their careers, you know, when they’re from the, the first day as, as, as a professional software developer or whatever they’re doing, product manager, designer and so on to where they are right now. And they have a level of anxiety that comes with that, you know, in the early, when you join a job, you’re worried about what is everybody thinking about you, right? Like you’re joining a team of people and whether you’re the most experienced person in the world, or you’re just coming out of college, you’re

you’re still thinking about how everybody else is thinking about you. That’s, think, part of the human nature. And I’ve been, you know, when I talk about AI and copilot, the nice thing about asking questions to copilot or ChatGPT or any other AI tools is that it doesn’t judge you. You can ask it, you know, an infinite number of questions and you know, it will not say, Thomas, you’re stupid. And like, you keep asking that same question over and over again. And then we ask questions, you know, to our teams, our leaders, our managers, peers, and so on.

Even our life partners, we do moderate that to a certain degree of, I see you’re getting annoyed by the questions I’m asking you, even though for me, feels like it’s good learning. So I think this number one is like, don’t worry too much about how you’re perceived. think most people actually on the other side are not judging as much as you think they do. And certainly leverage AI in these times to mitigate that.

If you just want to ask the AI of how to ask nicely a question that has already been asked and you don’t know how to explain that. But you know, I think, you know, everybody has their set of fears and I would give my past self the advice of you’re worried too much and you’ll sleep better without worrying that much. Whether that’s early in career or whether that’s being the GitHub CEO and you worry about how, you know, perceived on a podcast or you worry about the GitHub universe keynote, which is, know, at the time of this recording coming up.

Luckily not for me this year as a speaker, so I have a chill time leading up to Universe but I certainly will dial in and see what the news is and how the presentation is. And I will not see all the stumbles that the speaker notices when they are speaking and thinking, oh, damn, you know, I mispronounced this word or I skipped over a line in the script or whatnot. So I think this is like the best advice you can always give your past self is don’t worry. Everything is going to be fine. And you’re actually…

Most of these days they thought they’re really bad and that they’re turning out really great

Erika (17:35) I feel like that’s the flip side of caring so much about what we do is the side of maybe like caring too much and then caring about things that we maybe need to let go of like, yeah, the things that are.

Thomas Dohmke (17:40) Yeah.

Yeah. I mean, how often,

you know, do you go into the movies and then you’re worry opening slack when you come out of the movie theater, especially when you are in a different time zone away from, you know, the U.S. time zones, the Europeans and the Asians and like, hopefully nothing bad has happened. And then you see the incident command channel being lit up and you’re like, do I want to tap on it? Is just the regular maintenance or is it an outage and…

Even worse, right? If you get three pings from your manager and whatnot, it’s like, this is the nature of having real, you know, big platforms. There’s always something going on. And I think you got to live with that, with that mindset.

Erika (18:22) Well, I know you’re a big reader. On your GitHub, you even have, I don’t know how up to date it is, but books that you’ve read recently. So what books or mental models have changed the way that you think about engineering?

Thomas Dohmke (18:28) Yeah.

Yeah, I don’t know if this thing on the GitHub page, it might be a little bit outdated because I’m reading like multiple things in parallel and I think it only fetches whatever the last one comes back from the Goodreads API. It’s cool in itself that that actually works and it uses GitHub actions, which you all worked on to pull that information and update a gist of it. There’s actually, I have it somewhere, there’s actually a book on mental models, which I found really useful on leadership, not so much on engineering.

It has a lot of, you know, things that I found useful. Often, you know, I think the best books are the ones that you can just open a chapter and read it completely out of sync. Certainly nonfiction books and yeah, nonfiction books and fiction books. don’t want to have that. But that’s, that’s one. I really liked the, now you’re making me remember names of books.

I think it’s the engineering primer or something from the author that used to work at Stripe. What’s it called? Will Larson Will Larson has like a three, I think three books here. like the engineering primer, the engineer executors primer, and then there’s a third one. They’re all kind of like bit of different phases of your engineering career. And I enjoyed reading all three of them, even as I was already a manager at that point in time, because you learn a lot about kind of like advice that

Will is giving to engineers at different stages in career. It’s also useful to be as a manager. I think at the time I was special projects and then when I was CEO I read the engineering executives primer, actually with all of GitHub’s engineering leadership team. And it had a lot of useful advice about being an engineering executive and about how you handle, you know, your engineering team and culture, but also how you handle, you know, budget requests and how you.

understand the perspective of the finance team when you give them those requests, know, asking for my headcount. I made air quotes again, asking for my headcount is a very common thing in large companies. And it’s not always the right approach, but often, you know, it is a question of can I get more funding because the work that I have now on the platform that we’re dealing with, or the, you know, the billing API requests that we’re getting is so much bigger than it was when that team was originally founded So I found those.

I think it’s three books. We can add to the show notes the exact titles later.

Erika (20:43) I will add those to the list of must reads.

Thomas Dohmke (20:44) Hm. Hm.

all coding books, honestly. think it’s still even, you know, when you have 30 years in your career, it’s cool to just open a coding book on a new technology and read through it. And you don’t have to read books end to end. I’m a big proponent of stop reading a book when you get enough out of it and move to the next book, which there are so many books. like there’s so many coding books these days, you know, about building AI systems and understanding large language models and those things. I’m always excited about picking up a new one.

Erika (21:15) Absolutely. Aside from reading, what helps you learn something quickly and deeply? What are kind of your go-to approaches for not only understanding the surface level, ⁓ but like staying up to date?

Thomas Dohmke (21:28) Yeah.

You know, I think the first thing is to be constantly reminded that you’ve got to read something. So I’ve stacked, I have books stacked everywhere in the house, um, on desks and side tables and on my nightstand. And it’s less about that those are the books I actually need to read. It’s more about, oh, okay, I should read something. Um, um, and, it kind of like reminding yourself, um, that there’s something else to do because there’s so much.

you know, on our mobile phones and TV and whatnot to watch that really you’re limited in attention and the time that you have during your day, not in the information available. I’ve subscribed, you know, to a bunch of blogs. I still have like old school RSS reader, Unread on iPhone and on Mac. And so I go through those you know, on a regular basis, often during a morning coffee and then during the day when there’s a little break and what have you.

I have lots of browser tabs open and then I organize them into reading lists and then those reading lists become stale in itself. I think I read on some blog that most to-do lists are 95 % or so, some number of them are created and then never checked off. And it’s really the practice of creating a to-do list that gives you the dopamine hit. then while checking off these items is something you’re procrastinating forever.

So I think this is what was statistics from Evernote or something like that. I think the most important thing is really just what is the ability to read through a newsletter or blog post and quickly capture whether this is actually useful or not. And because there’s so much content and there’s often so long, if you have the habit, I need to read everything to the last line and not being willing to scroll through stuff.

You’re never going to have all the information that you need to get to the job done. So that’s the skill is reading fast is great. By the way, I love listening to podcasts at 1.7 speed. I also speak fast. So maybe that doesn’t work for this podcast. like, and then because you can get so much done while you’re going grocery shopping or mowing the lawn or driving to work or wherever you have time to listen to podcasts and audio books and what have you, because that’s a way of capturing information.

Otherwise you’re driving and you can’t really read a book.

Brittany Ellich (23:32) Yeah, that makes a lot of sense. I feel a little bit called out by that to-do list statistic because I definitely have a lot of to-do lists that have been created and not checked off.

Thomas Dohmke (23:41) Yeah, yeah. It’s, you know,

urgent things I need to do this week. And then you look at this three weeks from now and you’re like, oh shit, I didn’t do any of these things, but they were urgent.

Brittany Ellich (23:47) ⁓ no. Uh-huh.

Exactly. Yep. So I’m curious, we want to move a little bit towards what the future holds. And I know we can’t necessarily, it sounds like a lot of that is still under wraps, but I’m curious, what are you most excited about now that’s happening in the industry and how has that changed recently?

Thomas Dohmke (24:08) What’s happening is AI. And I think we’re in this phase where, you know, a lot of the companies out there that are part of our professional life, software developers, but also our personal lives, know, our banks, insurance, and whatnot, they were all built in an era before AI. And now we have, you know, a huge number of startups in Silicon Valley and around the world building for the era post AI and, know, chat GPT or so was that moment that kind of defines the before and after.

But there’s so much, I think, white space of reinventing a lot of the things that we do in life, that the opportunity both for existing companies to look at their business models and what the value that they provide to customers and new companies is to disrupt that old world. And we have seen these changes before. We already mentioned a few. We move from punch cards and…

mainframes to the personal computer. And the personal computer really was revolutionary because the job of a developer before the personal computer was to write something often on paper or so, wait for their slot on the mainframe. Then they had basically one-on-one time with the mainframe. They could run their thing, get the results. if you look at these mainframes, it was actually like the flickering of an LED or something was the output.

And then go back and then diagnose the problem. Like imagine that’s how we would do software development today. We would all quit our jobs and work on a farm or something. And then the personal computer changed all that. like same switch and all of a sudden, and you saw the early use cases for personal computers were actually spreadsheets because it was so much easier for an accountant or somebody that had to add up numbers to do that in a spreadsheet, even though there wasn’t a graphical user interface, then it was doing that on paper.

Because you can quickly, you know, edit a cell and it updates the sum, right? Like it’s so simple in today’s terms and it was so revolutionary when that started to happen in the 80s. Then we had the before and after internet and from being isolated in your home and trying to learn from books and that’s all cool until you get stuck. And then you only have the book available and you know, it’s 9 PM or whatever, and you can’t go to the library, let alone that the library doesn’t have coding books on computers that are only released in United States or what have you.

you had nowhere to go. the best scenario was that you could find another nerd in computer club or in school or in your network. I mean, literally you could send a letter to the magazine and hope that they would answer the next next month’s edition. Right. and then the internet came and with that, you know, use net and newsgroups and forums and, and, and, ultimately source forge and then get have their developers could come together and discuss problems. And

That shows how much that changed our lives as remote first company as we have them today as GitHub is and many of GitHub’s competitors, where all you know your coworkers from is a Zoom call, a Slack channel, and of course GitHub itself. That wasn’t possible before the internet. And I think that same change is going to happen in our life through AI. And it happened through mobile and cloud.

podcast has a certain limited length and then we have to stop it. But like we’re going to see the change, same change to AI again. And we don’t know what the future will be because we didn’t know that before the internet either or before mobile or before the cloud.

Brittany Ellich (27:08) Yeah, that makes a lot of sense. I feel like we’re all feeling that as well, trying to keep up with all the things that are changing right now and embrace it and see what’s working, what’s not working. It’s a very crazy time, but you’re right that this has happened many times before.

Thomas Dohmke (27:14) Yeah.

I mean,

classic example in software development, right, with copilot is when it started was on an auto-completion and that was already revolutionary in itself. The statistics of 55 % faster came from just auto-completion. Just by predicting what the next few lines are, you prevent the developer from going from their editor like VS Code over to the browser. Maybe we all have the rest of our life open in tabs and trying to find the answer to something that is predictable.

by looking at the code that you wrote before. Then came chat. All of a sudden, you could ask all the questions and it would actually understand the code that you have already open. Now with agent mode and coding agent, you can go even a step further and have it run for minutes to hours on end to solve a problem for you. That shows the paradigm shift that we’re going through.

Brittany Ellich (28:06) Yeah, I definitely have been spoiled by that. The other day I had to write tests by hand. was, I think I was on a plane, so I didn’t, and I didn’t want to spend the $8 on the wifi and like actually writing tests by hand, which I have not done in a very long time, was very painful.

Thomas Dohmke (28:11) Yeah.

Yeah.

should have spent the $8 and expended them.

Brittany Ellich (28:21) I know,

it would have been worth it for sure. And then what is the future for you, if you can say anything to her? Are you headed back to Germany? Are you, you do have plans on making this stealth startup an unstealthy startup?

Thomas Dohmke (28:35) definitely have plans to make the start up a not stealthy startup in the, in the, know, next three to six months, probably I’m going to, you know, stay in the US I love travel. And so obviously my, my home country, my parents and in-laws and, whatnot are in Germany, but I’m going to travel back and forth. But look, the future is, I mean, already talked about a lot about AI and we’re trying to see, you know, what can we change with AI in the life of a software developer? That’s the space, you know,

love to be in. I’ve been in for the last 10 years at Microsoft at GitHub. And I think there’s so much work to do and so many things to be reinvented. As we went through these positions before, nobody knows what that will look like. And I think those predicting even the next year are going to be wrong because if you look where we were with AI code generation, whether it was copilot, cursor, Claude code, it wasn’t even a thing a year ago, how much that has changed, how we’re thinking about the space.

I wouldn’t want to predict where things are in a year, but I want to be in that space and I want to work on these things. to some degree reinvent the software development life cycle, in other parts contribute back to it and really just enjoy seeing developers using my tool and sharing the passion that I have for it.

Brittany Ellich (29:46) Awesome. Well, with that, think we will start wrapping up and we always wrap up our podcast with a fun segment. Again, the air quotes, sorry. But what we’ll do is we’ll ask a question and we’ll do a round robin answering and you’ll get to go last so you have the most time to think about it. And this question came from Bethany. Fantastic, fantastic idea. if I know that you are a big Lego fan.

Thomas Dohmke (29:57) Yeah. Yeah.

Okay.

Brittany Ellich (30:08) And the question is, if you could will any LEGO kit into existence, what would it be? ⁓ And it’s a, yeah. Yeah. And it’s okay if it already exists because there’s a lot of them that, you know, there’s so many. So yeah, I’ll go first. So if I were to invent any LEGO kit, I think it would probably be, this probably already exists, but my…

Thomas Dohmke (30:14) And you go first. no, I want to hear the answers from you first. Yeah. Yeah.

Brittany Ellich (30:31) Kid is very into K-pop Demon Hunters and I’m sure this is coming out and she’s also getting very into Lego. So this is the perfect time to bring these together and I’m ready for the K-pop Demon Hunters Lego crossover because I’m also a big fan of it. That would be mine. I don’t know exactly, maybe their house. I think their house looks really cool and I would love to spend some time playing in that house with that very comfy couch. Bethany, do you want to go next?

Bethany (30:54) Yeah, I will say they exist in like a fortnight, which I’ve already done. For me, it would be, I really love the tuxedo cat and I think it’s so cute. It would be really cool if you could just do that for any of the pets that are close to you in your life. So like I’m thinking my parents’ dog George or my childhood dog Daisy.

Thomas Dohmke (30:57) You

Yeah.

Bethany (31:15) Lego kits of them would be so fun or at least like more options if that’s not already a thing.

Erika (31:21) I am a big fan of like the sort of like geographical Lego situations where like you go and there’s like a city built out of Legos. And I’m currently living in San Diego. So I think it’d be really fun to have like a giant San Diego Lego kit where you can build all the all the things and have little cars going around and stuff.

Thomas Dohmke (31:42) Well, you know, it’s actually a fun question, not an air-quote fun question. I have three answers. One is, you know, I love really old Lego, like the Lego that was being sold when I was a, you know, 10 year old, 12 year old, especially as I think most of us go through childhoods where you look into the Lego catalog and you see something you really want and your parents are like, it’s too expensive or Thomas, already have a train. You don’t need a Lego train if you have, you know, trains. And so as, you know, in the early nineties, what I really wanted to have is the

space monorail, which was just so cool. so a few years ago, I went onto Bricklink, which is a site where you can buy and sell both new Lego parts and also old Lego. And it’s like all these sites, right? know, the older it is and the more pristine minted box or whatever, the more expensive the set is. But I did buy myself that old monorail and it is in my shelf in…

And I can see it all day and being reminded of what I wanted as a kid and what I have now. So that’s, you know, that I think is if you really like old stuff, I can recommend Bricklink. It’s often cheaper and more nerdy to buy there than buying on eBay. I think the second one is like a more current thing. It’s just like the new Lego Death Star came out and it’s not actually a sphere, but it’s like a slice of the Death Star. And it’s very controversial because it’s thousand dollar set and doesn’t

doesn’t really show the death star it shows like a play interior, it’s like a puppet house. So I’d rather have a real death star if I could will that into, I think most nerds want, most Lego fans want the outside looking like a real death star And then the third one is, you all mentioned this is maybe the thing you really want is an AI system where you can describe what you want and it is good enough to assemble the bricks.

and then get to the part list and then Bethany, you can get your cat or your dog or whatever Lego. I mean, I’m sure the technology is there already. Somebody just needs to build it. And so maybe you need to will that into existence even better if it’s an open source project in GitHub.

Brittany Ellich (33:36) That is a really incredible idea, and I really hope somebody starts that. And if you do, reach out to us and share it, because that would be awesome.

Thomas Dohmke (33:41) Yeah. Maybe

you can use copilot to code it and then use AI with GitHub models or so to, yeah. It’s not my next startup, so don’t have to cut it out. But it’s a cool, I think it’s those things often hobby projects because the commercial value of something like that is only there if you’re Lego and maybe you sell it to Lego, but it’s not.

Brittany Ellich (33:48) That’s true. We can cut this part out if this is, maybe this can be your next startup right there. ⁓ Okay. It’s a cool idea.

Thomas Dohmke (34:08) startup idea in itself.

Brittany Ellich (34:10) sense. Well thank you so much for joining us Thomas, this has been awesome. If folks want to find you on the internet, where do you recommend they go?

Thomas Dohmke (34:18) Of course, I still have my GitHub profile, Ashtom. I’m also Ashtom on X and I’m Ashtom on LinkedIn.

Brittany Ellich (34:23) Great, well thank you so much for tuning in to Overcommitted. If you like what you hear, please subscribe or do whatever it is you’re supposed to do on the podcast app of your choice. Check us out on Blue Sky and share with your friends. Until next week, goodbye.

Thomas Dohmke (34:36) Thank you so much, bye bye everybody.