How He Built a $110M AI Company from Scratch (Dylan Fox, Assembly AI)

**Dylan Fox** (0:00)
I feel like some of the most challenging times at our company came from blindly following startup dogma or conventional wisdom. And that just didn't work for us because our customers don't want to buy that way. A vision I have for the company is like, I want us to just be like insanely customer focused, like insane. Our goal is to build the most accurate, the most comprehensive, and the easiest to use developer platform to build these speech and voice AI applications.

**Wouter Teunissen** (0:25)
How do you deal with that responsibility, specifically as a founder, when you keep raising more capital?

**Dylan Fox** (0:29)
I think we went from 40 people to 115 people in under a year. It's way too fast and definitely wouldn't do that again.

**Wouter Teunissen** (0:36)
It's interesting. Do you then generally think hiring for someone to figure it out, as opposed to hiring for someone to solve?

**Dylan Fox** (0:42)
I don't think you should ever hire someone to figure it out. There's this image of what it's like to be a founder, and then there's the reality. For some reason, more so now, I think you tell me what you think because you've consumed this content for a while, but I feel like there was like a productivity porn or whatever. There's this image that I think is unhealthy of what a founder or what an entrepreneur is like, because you try to aspire to something that's not real, but then there's the reality. Once you realize, oh, this is actually what it is like, then you can, it's more fun.
I've found myself asking myself this question of, where are these expectations coming from, of how this thing should be or whatever? And it's like, yeah, actually, I don't know. It's like, maybe this is just what it's like. And then you can just lean into that.

**Wouter Teunissen** (1:41)
Yeah, that's interesting. I mean, I think it's like Charlie Munger says, lean into a natural drift. There are a thousand ways to build a company, and I think you're right. Like, I obviously consume a lot of this content too. And it's, I think there is like ambition, more like, you have to do it this certain way.

**Dylan Fox** (1:54)
Yeah.

**Wouter Teunissen** (1:55)
And like, that just might not work for every single type of company or type of founder.

**Dylan Fox** (1:57)
I call this like the startup dogma.

**Wouter Teunissen** (1:59)
Okay, yeah.

**Dylan Fox** (2:00)
And I actually just wrote a message to the company. It was like, beware of the dogma, because I feel like the worst times at our company came from like following the dogma.

**Wouter Teunissen** (2:10)
Interesting.

**Dylan Fox** (2:12)
And it's like, yeah, when you're first time founder and you don't know what you're doing, like you do, are we going by the way?

**Wouter Teunissen** (2:18)
Yeah, I think we're recording, right? Are we recording? Yeah. Okay, I mean, I don't like doing the intros anyway, because it feels more natural this way.

**Dylan Fox** (2:26)
Yeah, well, it doesn't matter. So I'll just keep going.

**Wouter Teunissen** (2:28)
Yeah, keep going. This is on a roll. I'll probably at one point bring it back to the early stuff. Because you keep talking, because this is interesting.

**Dylan Fox** (2:33)
Yeah, so I feel like, I lost my train of thought, but like there's-

**Wouter Teunissen** (2:37)
You sent the memo.

**Dylan Fox** (2:38)
Oh, yeah, about like, beware of the dogma, because yeah, our hardest parts, I feel like some of the most challenging times at our company came from just like blindly following like startup dogma or conventional wisdom.
Yeah, it's like when you're a first-time founder like me, or like a lot of founders, even if you're not a first-time founder, but it's your first time at a certain stage or whatever, you don't know what to do all the time. And so you follow the dogma and the conventional wisdom, but that can just get you in so much trouble. Because something I think about, and you just said it, but it's like a startup is not a franchise business. So if you're opening like a Denny's, I think Denny's is franchised, right? If you're opening a Denny's like, yeah, it's like, here's your brand kit, here's your operating procedure, here's how to like make the pancake, you know, it's like the exact amount of time per side, right? Don't deviate, like this is the recipe for success. And it probably works, that's why it's like all templatized. But I think with startups, it's like, what I've learned is like every market is different, every customer is different, every founder is different, team is different. And so, like, like, of course, there's no template to follow. And you have to figure out this thing, I think of now as like company market fit. So it's like, what's the best way? Now what I think about is like, what's the best way to run the company that enables us to best serve our customers? Like, not just what's the best product to build, but like, how do we do roadmap planning? How do we meet internally? How do we, like everything? What metrics to track? It should all be in service of, does this help us better serve our customers? If yes, then like awesome. If no, then like, why the fuck are we doing it? And like, you know, it's probably because of some expectation on like, startup dogma or whatever. Right. So yeah, I think it's like, it's something I think a lot about now.