Building with LLMs, Finding a Co-founder & Other Listener Questions

0:00Welcome to Rework, a podcast by 37signals about the better way to work and run your business. I'm Kimberly Rhodes with Jason Fried and David Heinemeier Hansson, co-founders of 37signals. This week, we are knocking out some listener questions. We got some video questions in, which we love to receive. The first one is from Ryan.

0:17Kimberly Rhodes:This is Ryan Chitwood. I'm an independent iOS app developer on the side. And since I work by myself, I've been wondering about adapting your setup of one designer and one developer comprising a team. What do you think about these days with the advent of LLN tools like CloudCode? Now, these are getting pretty good at coding at this point. What do you think about teams of one, that being a designer with some development skills, using a CloudCode or something like that as filling in the developer portion there? I don't know if y 'all are ever thinking about moving to that model at 37signals, but what do you think about for an independent person like me?

0:52Kimberly Rhodes:Thanks. David, I'm going to start with you. And then Jason, I'm sure you have some things to add on the designer side. I think this is one of the most amazing things about the LLMs. And it's something we've seen already at 37Signals in a big way. JC in particular have been really good on our team of just running ahead of programmers. And he's able to produce so much more, so much faster with the LLMs than he was before. Now, it all depends on what kind of fidelity you need. At our scale, the number of customers we have and the criticality that we're dealing with, We can't usually ship a bunch of LLM code into production.

1:31It's still just not good enough for that yet. There's obvious errors, obvious mistakes. There's barely a day that goes by right now without someone getting exposed online for shipping some LLM slot that then left a huge security hole. There was just this whole bit about tea. I don't know if you saw this, this app where women could talk about their experiences with men. And apparently it was done by this guy who literally had six months of experience as a programmer on a boot camp. And I'm sure he used a bunch of LLMs and everything was basically just wide open. So here you had a runaway success app that spilled the tea on two million women because it basically wasn't locked down.

2:14And that's, of course, sort of the nightmare scenario. Not that these things can happen with quote unquote real programmers, human programmers, whatever you want to call it. but it's far less likely that it's going to happen. So I would look at what kind of apps are you working on? Are you working on things where if user data got leaked, it's potentially a major, major problem? Then maybe I would think about having someone more skilled come in and review it at least at the end. But if you're working on, say, a game or something just fun and frivolous, maybe something that doesn't collect user data at all, who cares?

2:49You see LLMs all the way, get some experience, get something out there, see your ideas in action. I mean, half the enjoyment with LLMs is not necessarily that you're going to ship it, but that you get to see your idea actually work. And half the time when you see your idea actually work, you go like, well, that's not quite what I meant. Bam, in trash it goes. And then you build something else. So I do think it's going to be more and more feasible for single teams. I mean, I've experienced the same thing on the other side, that putting a quick design together. All the first Umachi, Omacube, and Omacon websites were all almost one-shot LLM productions.

3:32Like, they were kind of a little bit janky, but good enough. And I just hit launch. And then I had Sean, one of the designers we have here at 30s, come in afterwards, spruce it up, tune it up, make it better, make it nicer. But the ability to just ship something and see if there's something there is incredible. I mean, I absolutely love it. So I would feel no shame about it. Again, unless you're actually dealing with personal information that can't stand getting leaked, then you should do it. Jason, anything to add?

4:01Kimberly Rhodes:Not a whole lot. I mean, I think it's great to be able to get quick momentum on an idea and see something come to life. My question with all this stuff is always that like, then what? Now what? You can get yourself in a position where you're overextended too far out. you built this thing, you don't really understand how it works. You're the only one around. And now something goes wrong and you ask the LLM like to fix it. And it can't because it just can't for whatever reason, or it's fixing something. It's not really the problem. And it creates another problem. And now you're like in this place where you're, you're over the cliff already.

4:32Kimberly Rhodes:And you're going to like fall and you don't know what to do. And you freak out. That's like a dangerous place to be. I think with anything is, is beyond your own competence and your inability to deal with reality at that point. So I think that's the threshold for me is how far can you take this to what degree? And then are you comfortable like putting this out there for real people to use? And what happens if something goes wrong? Because things will go wrong. Software has bugs, all of it does. And when you don't know how to fix it, it's a real problem. So that's it for me, but I agree with the criticality is also obviously a huge factor, primary factor here.

5:07Kimberly Rhodes:But still, even if it's not that critical and something breaks, and you don't know why, you don't know how, you can't fix it. And this happens with LLMs, they get sort of in this loop where you can't quite get to the thing that's actually broken, or you don't know how to ask it how to fix the thing that's broken because you don't even really know what's broken and you describe it in a certain way, but it doesn't quite do it, then what? So just keep in mind that that could happen. And maybe it's a good problem to have. I don't know, maybe it's not, but it's a problem you're going to have probably.

5:35Kimberly Rhodes:And if you don't know what you're doing, I can imagine it can really inject a whole bunch of anxiety in the project. Yeah, great feedback. Okay, the next question, also video question, is from Anton. Hi there, Anton here. I'm trying to learn more about starting a business and sampling knowledge from both aisles of the VC disputes. And so I did startup school and now I'm reading Rework. So my question to you is, how do you guys think about finding a co-founder? Y Combinator strongly recommends it, but, and you two also have a nice team, but is it possible to go as a solo entrepreneur? I mean, it's obviously possible.

6:12Kimberly Rhodes:A lot of companies are run by one person or founded by one person. I mean, we didn't find each other in that way. So we didn't like set out to find someone else to do something with. It just kind of evolved that way. This is my bias, right? Which is work with someone first, get a feel for who they are, what they do, how you work together, and then maybe something else can develop versus going out there and trying to find your wife. you know it's like you gotta date first right and so problem with business is that like people sign contracts and they you know this is i guess not that different from getting married in a sense but but no one would just like get married because they found someone at some conference i mean i guess you could do that but most people don't so so you got to find a way to to date through work in a sense and see how it plays out that's kind of my big thing i would say though that to me this idea that you must find a co-founder i don't understand that i don't i mean i get it from like YC's point of view and VC's point of view, which is like, well, if one person leaves, like we haven't lost the whole damn thing because that's a huge risk for solo founders and someone who's investing in a solo founder is if that person dies, leaves, quits, whatever, the whole thing is like could be over tomorrow in a sense.

7:19Kimberly Rhodes:So I get why they actually want to back up in that case. But I guess if I'm going to shift gears just for a second, if you are going to look for a co-founder, I think you want to find someone who does something different than what you do. You probably don't want another technical co-founder or designer co-founder necessarily. You probably want someone who could do something else. There's a big overlap on maybe the business side of things or general values and philosophy. But make sure if you have two people that you're getting two people's worth of stuff and more surface area coverage versus just trying to find like a clone of yourself to kind of do what you already do.

7:49Kimberly Rhodes:I don't think that's really valuable. And it probably is going to end up, you're going to end up hating yourself in that case because you're going to butt heads all the time over the same kind of work. And I think that's going to be a problem. I think what's interesting about this question, where do I find a co-founder, is that it's often premised on like, I'm sitting here at home. Like, where are they? Why aren't they showing up? Where are you showing up? Where are you going out? And what I've seen in the Danish startups that I've invested in is that a large percentage of them have co-founders who met in college.

8:20And that alone seems to be a great reason to go to business school, to go to program school, go somewhere where you're going to meet other people. And if it's not going to be that, then you've got to involve yourself in some communities where you're likely to see that. I think open source is another option, especially if you're working on something deeply technical. And in that scenario, you might actually have multiple technical founders, but you've got to get out in some of that. This question of, I'm just going to meet someone is, in my opinion, delusional. That just doesn't happen. You have to put yourself in the kind of situations where you would meet someone.

8:55And I think college is actually one of the best options that you have. When I went through Copenhagen Business School, I had several people in that class that I worked with. Well, I could imagine we could have started a business together. I then ended up meeting Jason over the internet, and that worked out pretty well. But I think that's probably the minority case. The majority case is you're going to find someone more local to you that you actually get a chance to spend some substantial time with. That it isn't just this speed dating encounter. You've got to know someone for several months before you know it.

9:29Because I've also seen plenty of times that someone do get started with a co-founder and it turns into shit. It turns out terribly because the people are not aligned, they're not compatible. I guess now we're into the parallels of marriage there too. that's going to happen sometimes. Not aligned, not parallel, all these other things. But there's no silver bullet. There's no magic recipe where you're going to, you've got to get around other people for substantial amounts of time. It is quite unlikely that someone's just going to fall into your lap, either in business or in marriage. Okay. Question, I'm going to read this one to you guys.

10:05You can tell me you don't want to answer it because we've had a lot of people ask similar things in the past. This one is a YouTube comment. It says, do you guys ever think about the next generation? And by that, I mean, what happens when both of you decide to step away for good? I don't need a business plan, but do you guys ever think about the transition or how to find the people to fill your shoes?

10:25Kimberly Rhodes:I don't think we would find people to fill our shoes. I think if we decide that we're done or when we decide we're done, because at some point that'll happen, I think we'll sell the business and that'll be that. I mean, anything could happen, but it seems unlikely that we would transition the business. There's a variety of reasons why. Part of it is the way our ownership structure is set up that like David and I would still have to be involved. And that doesn't work. I don't think you can have founders involved in a business on the sidelines. I think if the structure was totally different where, you know, the majority of our income didn't come from profit distributions and whatnot, it might be a different story.

11:00Kimberly Rhodes:But the way we're set up, that's just the way it is. And I think it would be unfair to another team, frankly, to have us anywhere in the ballpark around because I don't think it would work very well. I'm of two minds here. There's something nice about continuity beyond the original founders. I think there's also something really beautiful about companies that are run by founders. And when the founders aren't around, the company is over. I think things should end at some point. It's like a band. You don't replace the band. I mean, you can, I guess, replace the band members. You have three people in a band or two people or five people.

11:35Kimberly Rhodes:Sometimes someone comes in and replaces the singer because they die or the guitarist who gets stoned and doesn't show up anymore for practice. For the most part, like the founding group is either there or they're not. And if they're not, it's not the same band. And it would have been better off had the band just stopped. So I think ultimately that's probably how this will play out. But I don't know. And we're not for sale at the moment. So I can't say for sure. But that's what goes through my head, at least, honestly. I think what really brings me to Jason's conclusion is the fact that we've gotten some trial runs.

12:04Both Jason and I have been involved with other businesses, including things that we had spun out from our own business along the years. And what I realized is I am uniquely ill-suited to sit in the passenger seat of any car that I feel I have ownership of. If I'm not driving, I'd rather drive that fucking thing into a wall. And that's a character flaw. Like, I prefer it to be otherwise. I'd prefer it to be able to just sit patiently in the passenger side and just offer sort of little nuggets of advice. And then I don't care where we're going. Like, that'll all turn out. Like, that's just not who I am.

12:41And I think that is the reality of a lot of founders of our size businesses. because I do think that when you get to a certain size, certainly by the time you're a public company, it is different. Not that they all pan out well once the founders leave. In fact, there are endless stories about how they don't pan out well. I think both Howard Schultz at Starbucks and Michael Dell at Dell tried to hand over the reins to someone else, realized they didn't like at all where that person took it and came back to it, even though those are two of the biggest companies in America. So can you think the risk is there even at that scale.

13:15But there are also clearly, of course, other companies that have managed to create that transition. At 60 people, 40 people, 80 people, I think it's a lot rarer that the ship can just keep sailing in the way where someone is replaced, the founder is replaced by someone else who just happens to have all the same qualities and urgency and judgment and taste and whatever that made the business work originally. They just slot in and it just all works. I actually can't think of a single example off the top of my head. I can't think of a fair number of examples where those kind of businesses end up getting rolled up or bought out by someone else.

13:53And in some cases, it even works. Fair number of cases also doesn't. But it seems like that's the main realistic door out. What I certainly don't want is when the day comes and the station says it will at some point where we no longer run the company, I don't want to be associated with it at all. like whatever it's got to be a million miles away and if the next owners of the business choose to do something that i could not tolerate the whiff of if i was in the car i can just go like well ain't my business anymore and then that's that so yeah i think we're very much on the same page about that it's got to be someone else's i was about to say headache that's not even true this isn't a headache, at least not most days.

14:39This is a pleasure. This is a privilege. But at some point, the privilege is going to end and then it's got to be a clean separation. So no freed Heinemeyer Hansen children taking over. Do you know what? It's funny. I was just thinking about that when I was saying I haven't seen it pan out well. I do actually think family-run businesses do pan out well. That's exactly what I was thinking. Yeah. The economy is full of them. And in some countries, it's even a majority way of running the middle stand kind of businesses. Germany. Yeah. But I also think that I don't know if I've seen it in tech. I actually wish it was different.

15:11Jason, you made the point the other day about what if we were making knives? I can totally imagine the son of a knife maker becoming a knife maker. I have a little bit of a hard time imagining my children sort of wanting to do exactly what it is that I do with computers.

15:25Kimberly Rhodes:Same. Yeah. Yeah, that's an interesting thought. I haven't seen that. Have there been family-run software businesses that have transitioned between generations? That's an interesting question, actually. I mean, there haven't been a lot of software companies that have been around for generations. Well, that's true, too. But there have been a number that have been around for 50 plus years. I don't know. That's interesting. But yeah, I don't see it either. OK, let's go to this question. This came as a text from Brent. It says, I'm a longtime fan of Ruby on Rails, but only recently started listening to the podcast.

15:55I never realized that 37 Signals has produced so many products over the years. Question for Jason and David. How do you come up with new product ideas and validate them before investing a lot of time in building them? I know you two are not fans of MVPs. Do you see this ability to find new product ideas as being an innate gifting or something that can be cultivated? So kind of two part question. How do you come up with the new ideas? And is that just a natural talent that you guys have?

16:24Kimberly Rhodes:well just to be very humble about it like we haven't had a lot of huge hits so we we've had base camp it's been a massive massive hit high rise was another big hit hay has been a a good hit but we've done a dozen things you know so like our record we have a decent record luckily we have one massive massive hit which is base camp but you know we can shake ideas of our sleeve but most of them don't work and i think that's just the reality anyway but i think it's fun to just keep making new ideas and trying new things occasionally. And some things come from like direct needs that we have. Primarily, it's that.

16:59Kimberly Rhodes:It's like, we need this thing, or we have this idea for something we want to use that we don't have. Like, we're talking about a new product right now. We're building something called Fizzy, which is a bug thing, and we're building a new version of Basecamp. But there's another idea that David and I have been chatting about a little bit related to some stuff that he's been going through with the Linux stuff he's doing and this community building. And like, God, that would be really nice to have for our products. And so like there's a spark there, right? It's not this, we have to drop everything kind of spark, but there's a spark there.

17:30Kimberly Rhodes:And it tends to like, does it ignite a fire? Does it ignite something? Is there heat there? And then sometimes you just begin to make it. You don't put everything aside. You make it on the side and you see where it goes. So to this point about this question about validation, we don't validate things externally in that way. I don't believe there's such a thing, frankly, because I think you can ask people all sorts of questions and it doesn't matter because they're not real answers. Like, would you pay for X, Y, or Z? Sure. Well, if you don't have to pay for it, it's not actually an answer. Or would you use X, Y, or Z if it did this?

18:04Kimberly Rhodes:People are like, yeah, sure, I'd use that, but that doesn't mean anything because I can't use it right now. So I'm not a big fan of that. I'm a big fan of internal validation. Like if we're going to build something, are we going to use it ourselves? Are we using it? What does it feel like to use it? And we've had a few things that have started then died, and we've had a few things that have continued. And then we've had a few things that have continued and continued and continued and continued. But I don't think it's a particular talent, frankly, to come up to an idea. The hard part then is to form the idea and to see it through and to figure out if you're willing to bet on it.

18:35And if you think there's something there, those are the things that I think

18:38Kimberly Rhodes:you need ultimately to see something through. But I think a lot of people have ideas. And especially in this era, ideas are everywhere. It's like, which ones make sense? How are you going to guide that idea out? Where are you going to stop the idea? That's the other thing. An idea naturally has no boundaries, so it can get as big as you want it to be. Where are the boundaries? I think that's another thing. Or is it disciplined to say this is enough of this idea before we launch it to see how it works? Those are some of the thoughts that I have. I think part of the problem here is seeing it as an issue of talent.

19:09Do I have the talent to pick these ideas? I don't think it's so much about talent as it is about exposure. If you work in a company, if you work for other people, if you work on something that already is a business, you will literally be forced to run into a string of issues of stuff that's broken. You'll be forced to use all sorts of systems that are crap and you have all sorts of ideas of how to make them better because you have to use them for a specific purpose. It's the exposure to real problems in real domains that give you the ideas. Now, we're fortunate at Thersima Signals that we have a relatively large company of 60-odd people.

19:48You just run into problems all the time managing 60-some people, managing the products we already have. It kind of just feeds on itself. But I don't think anyone really has the talent that they could just sit down in a vacuum and come up with amazing ideas and then internally in their head actually decide and weigh them and see whether they're good or not. I don't think it works like that. I think you have to get exposed to the world. And the world is going to tell you, does this feel like something people would pay for? You're working for someone else. Would they pay for a better way to do this?

20:19They probably would if it would save you a bunch of time on some drudgery that you're doing anyway, or you're dropping a bunch of balls. You just get a vision of what good is when you're dealing with real problems. It's not something you can just cook in your little noodle by yourself. You have to actually get exposed to it. So less about talent, more about exposure. And this is one of the reasons I find that folks who are a little bit too much in a hurry to start a business. I mean, I was totally in a hurry, too, so I can very much relate to it. But why are you in such a hurry? Can you just expose yourself to a little experience first?

20:57I mean, now I really sound like an old person. And the reason I say it this way is actually, I think it's a Hemingway quote about great writers. that to be a great writer, yes, sure, you have to have some fluidity of language. You've got to have grammar down in some ways, at least so you know the rules, so you can break them. But then also, you have to live a little. You actually have to have had life experiences that are worth telling other people about. Very few individuals are just born with this magic ability to tell interesting stories if they haven't experienced anything in life. And I think the same thing is true in business.

21:34you actually get more ideas for business when you do business. And a great way of doing that is working for someone else.