Building Got Easy. This Startup Solves What to Build | Alfred Wahlforss, Listen Labs

23:30DCs use it for diligence. So you'll have this whole process of talking to the customers of different products and understand like, do they actually like it? So that's another use case. You can imagine Procter & Gamble, they're constantly launching new products in new markets. And to launch one of these products, it's tens of millions of dollars in ad spend and also like retail and shelf life. And if you can validate and understand like how you should launch it,

24:06Turner Novak:it can save a lot of money. Yeah, it's always interesting. Like you're Procter & Gamble and it's like, all right, we're coming up with like a new chocolate. Like do people like chocolate? Or like, should we add like, you know, dark chocolate? Should we make it 70 % instead of 60 %? And they like do this whole research campaign. People, they'll talk to this process of like hundreds of people to like make it like change some food or like change the packaging like it kind of seems a little bit silly i guess but there's just so much at stake that they definitely feel like all right if we make the package green instead of blue how will that change the perception and what's the roi on that so i guess it it sounds a little bit ridiculous but also like it makes total sense that especially the more resources you have the more you'd spend on this stuff.

24:50Yeah. And, and like, it can have a huge impact, right? The package that you choose. Um, and I mean, it can even be like, you, you constantly make decisions every day that in some way, like you're not fully aligned with your customers. You don't know exactly what your customer would want in one case, but I useless. And myself, we have created a simulation of our customer base. So the way we can talk about that, but we built this ability to interview one person and then create the digital twin of them by doing essentially a one hour long form interview. And then you can scale it up to a thousand people.

25:34So you have a representative sample. And the other day I was figuring out what's the title of my talk for a conference with our customer base. And it's like a really small decision, but it actually does matter. Like, are 20 people going to show up, 50 people going to show up? And by iterating with this like synthetic panel, I was able to get to a much better result than I initially had. And so I think that if you can help improve every single one of those small decisions, like you will have meaningful change in a large company.

26:14Turner Novak:And you said something interesting about this, these like synthetic sort of personas or data sets. How does that work and how is that useful? I'm just curious because I'm just like thinking, do you run into like different biases or like, you know, it's not actually real customer data cause it's like synthetic or made up. Like how does that actually work? Yeah. So our core product is really focused on talking to real humans. And we realized that we've done more than a million interviews in the platform now, and it's grown exponentially since we last reported it. And we said, like, what if we train digital twins based on all of those interviews?

26:58That would be really powerful. And you can think of this as like, if you have a partner, like you spend a lot of time then probably you can predict to some degree what they're gonna like

27:09Turner Novak:and not like like if they'll like a new food or if they'll like a movie or something like that you think you can do that uh like barely i will say my wife can probably do that much better about me than i could about her so i can give it like depending on what it is i could i could probably call it but she knows me like so well she'd be able to like anything like oh yeah turner would wouldn't like that. Okay, so your wife has a good model of you. But it turns out that LLMs can build this model quite successfully. So we have, in some cases, 95 % accuracy. And we measure that by just removing one of the questions from the training set and testing how well is AI able to predict the answer to this question.

27:56And so you can get very high accuracy. The problem is that obviously there are questions you can't predict, right? And so the model needs to know what it can answer, what it can't answer, and what's the confidence interval. What the use cases are, I would say is like brainstorming. The 99 % of decisions where it's like too difficult to talk to real people or you need answers really quickly, or it's like really a small decision, but it still matters, like the title of a talk. Or if there's hard to reach audiences, like high net worth individuals, really expensive to talk to. Now you can create these simulations of them.

28:39If you just talk to a hundred of them, you can like have some kind of simulation.

28:43Turner Novak:So like, I'm trying to think of like what a, what something can be. So like if I'm like Doritos, I mean, I feel like or like Taco Bell, like they always come up with these like crazy new products. So I can maybe say like, hey, should I make a strawberry flavored Dorito? And I could probably go into the listen data set and like a bunch of people maybe mentioned how they like strawberries or something and or they don't. So I'd be able to maybe get like a little bit of feedback on like, hey, it looks like people may actually be interested in strawberry flavored Doritos or you have enough history say, people probably won't like that.

29:18Yeah. Like one of the good best use cases, I think, is message testing. So that is basically what is the title of this billboard? What should I like? What should I name my product? These like really difficult, vague decisions that may or may not have some reference that you don't know about that went viral like a few weeks back and you'll be ridiculed by it like or you know a specific set of framing also even like aligning yourself and aligning like other people I've actually created a synthetic version of myself and sometimes when I have like decision fatigue I'll like throw that in like what should I have for lunch and I'll just let my synthetic AI choose for me because it's just easier to have someone else make the decision so like there's value in in getting faster to the positions one thing i maybe like relevant is like

30:16Turner Novak:for for this podcast when i'm trying to think of like what do i title this thing what should i put in the thumbnail on youtube i always just basically i copy and paste the transcript and i have a cloud skill that'll just basically like just bang out a bunch of ideas and like nine out of ten are pretty bad, but there's usually some in there that are pretty good. I'm like, oh, I did not think about framing it this way. I haven't thought yet what the title of this one could be. But even when I asked it before for prepping, part of it was like, oh, you should give it this immigrant to successful founder type of framing, or you should give this AI unlocks the qualitative side of humanity, even though it's like a very quantitative or something.

31:02Turner Novak:And again, it was like, I didn't feel like any of those really hit. So I'm going to see after this conversation, I'm literally going to throw it in, throw the transcript and be like, what are some ideas? But it always comes up with, usually there's a couple that are pretty good that I wasn't thinking of. Yeah. And what's interesting is that the taste of the models are trained on the average user. and so when we try this we asked ask claude chat tpt like what what do you think like even if you tell it like hey you should act as a market researcher whatever um it has different opinions than our synthetic or our digital twin panel and so it's not as aligned with your specific segment.

31:47So imagine if you had created a simulation of your user, like the people listening to this pod, you could then have that in as an MCP and let Claude kind of iterate together with that simulation to come up with the perfect title.

32:04Turner Novak:Interesting. I need to figure out a way to automate because it's all still kind of manual. I need to do probably some more like co-work automations of like when it notices that I've recorded an episode, it like will automatically go and run. I haven't gotten that far yet. I need to. I think this like begs it maybe an interesting question of if I'm a brand, like couldn't I just go to ChatGPT or Cloud and just be like, hey, here's what I'm thinking. Like, what do you think? Like what's the value of using something like Listen versus just a more general AI tool? Yeah. So the value for simulation is that the results are different.

32:41Like if you ask Claude, it has much worse taste than the simulation because it's not based on your specific sub-segment. If you think of something like Sweetgreen, you would think that, okay, that's a general audience, but actually it's high income, it's urban. And by the way, they need to know what seed oils are. And all of a sudden it's like a very small subset of the population. And it lacks a bunch of the nuance that you get from the interview. So we see a very meaningful lift in the accuracy. So when you look at pure Claude, accuracy is around like 40%. And we get 95 % accuracy in some cases.

33:27Turner Novak:What is that accuracy? Like that 40%, what is 40 %? And then what's that like 95 %? Is it like the success of an outcome? Yeah, it's the like mean average error in answering a question. So we will remove, you know, 10 questions from the training set and then which the training set is basically the interview where we interview someone for one hour and we'll remove a bunch of questions and then predict how will we answer this question. and we let Claude do that. And we have the real answer as well. And then we see like, what's the average error? And we get about like 5 % of error. I think one of the things, you mentioned it a little bit earlier, that you work with Sweetgreen.

34:16Turner Novak:I think it'd just be interesting. They actually developed a product using, listen, what was that? Like, what did Sweetgreen use you for? Yeah, so they came to us and said that the menu has had issues with protein. And we did a study where we interviewed sweetgreen customers and ran hundreds of interviews. And Lyssen came out with an idea that they should create a new bowl, should call like the max protein bowl. And they ended up actually launching that and became a huge viral hit. And a lot of people are buying it now. So that's the kind of use cases that work really well when you're trying to do ideation or concept testing and you see how people react to these new ideas.

35:09Turner Novak:Yeah, I think they I feel like Sweetgreen is really good at being on sort of the forefront of new technology that comes out. Actually, the very first guest of the podcast was Jonathan Neiman, the CEO of Sweetgreen. and I think it was like at the time they had just launched their robotic kitchen where they were like using these like autonomous robots to like automate some of the preparation so and I feel like they were pretty early on like mobile takeout like mobile ordering and like takeout which obviously you can optimize the kitchen I've looked at the stock price recently but I know that I feel like I just generally that category has been struggling a little bit just those like pricing consumers are getting a little bit upset about like the Chipotle slot bowl, slot bowl memes.

35:56Turner Novak:I'm sure you've seen those. Yeah, but they're kind of, they're really great at testing new things and they've been amazing like partner from the beginning. So I know in just generally AI, there's kind of this like Jevons paradox thing where the better it gets, the more that you do. Is there a similar element going on with broadly customer research? Are you finding that people are doing more and more, talking to their customers because you make it easier and faster? I think there are these examples where there's no limit to how much you can get, how much value you can get out of a specific segment or of a specific task.

36:38And customer research is one of those. You can always perfect whatever you do to make sure that it's fully aligned. And our vision is to create a world that finally works the way people want. And there's so many small things that are misaligned with what people want. So we actually see that now that you can launch something, like one of our customers, they used to do these things once a quarter. Now they do it every week. There's a new product or something or new like event of some kind? like customer research. Essentially, they used to work with one of these agencies once a quarter. And that means they can fundamentally launch more marketing campaigns, more products.

37:25They can iterate much faster. And their products are more aligned with actually what their users want.

37:32Turner Novak:You're basically just like tightening the feedback loops, speeding them up. They're able to talk. I mean, really talking to your customers. Like if you go back to what is YC, like the advice to like when you're starting your company, It just talks to your customers, build a product that they'll pay you for. That's basically what you're helping people do at the end of the day. I mean, what I'm really excited about is when the coding models get really good, the YC model is write code, talk to users. And I think the coding models are almost good enough for this. We can essentially give a listen amount of capital and then go and talk to users, figure out what they want and build it and run that in a loop.

38:12and you have an autonomous organization.

38:15Turner Novak:So that's pretty interesting. To what extent can you do that today? Are there certain points where it just doesn't quite work yet because the technology is not there yet? There is still the judgment of when to ask and when to build and the models, the reliability is not quite there yet on the coding. But I think towards the end of this year, there will be huge improvements. And especially with the simulation where you can get really quick feedback, I can see the way we develop will be quite different. So you can basically have somebody in product who is talking, using Listen. It's going out and talking to customers.

39:03Turner Novak:They're getting feedback on it. And then they're like, okay, Devin, just go make it. And within the course of the day, maybe there's time windows for all these things. But you're basically just kind of sitting there and you're talking to customers. And then there's almost this triangle of product, customers, engineering. It's all in one, maybe. Because today, the preference model is you as the builder. You're building it for yourself. And you kind of have to think, okay, what would our customers actually want? What do they actually care about? But imagine if you could have a simulation of your real user, and that's just going to be so much more powerful.

39:41Turner Novak:Is there people that are doing that well today? Do you feel like there's any companies that are the closest to that? Or maybe how do you guys do it? I don't think anyone has cracked that yet. I think we have an edge because we're talking to real people all the time. So we have this extremely rich data set that we can train on. And that's why I'm excited for this direction. and we're hoping to launch this in a couple of months. Oh, so it's not out yet? Yeah, it's not out yet. Oh, interesting. Okay. What's the challenges in building this? Like what's been the hardest part of actually making it practical and usable?

40:15Making it accurate. The models have a bunch of, you know, they're super smart, right? So they will sometimes act in a way that's not in tune with how humans work. And a bunch of issues around that, basically. That's the hardest part.

40:33Turner Novak:Interesting. Yeah. Yeah. Cause it's, it's, it's sort of like with anything like AI, it's like, how do you quantify everything? Like everything needs to be a data point in a sense, but this is still a very qualitative thing. Like how does something make someone feel? It's kind of like this weird balance of how, I don't know, how do you, how do you balance it? I don't know. I don't know if there's an answer, but. You will not be able to replace all of the work we do with simulation because there is something about talking to real humans and seeing them react in ways that are just impossible to predict.

41:11And also being able to share highlight reels of how people actually feel when they see your product. A big value of research is aligning people, motivating them to actually go and fix the problems. Because sometimes you know all the problems, it's just there's no one actually going and fixing them. But having real people react to how bad your experience is can be a really great catalyst to make that happen.

41:37Turner Novak:Interesting. Yeah. Cause I feel like we're, and maybe an example of that happening right now is like a lot of people are now starting to build products and software that's kind of agent first instead of human first. Right. And like a year ago, that probably wasn't necessary, but we've kind of, as more and more software moves to being more of agents interfacing with other agents. The human-first software is not quite built correctly or in the same way more efficiently. So it's like this new problem that emerges where probably like a year ago, nobody would have thought this was a thing. But then now as an industry shifts, as behavior shift, demand use cases shift, all of a sudden it's like, there's actually a need for this to exist.

42:21Turner Novak:And it wasn't there six months ago. Yeah. But those agents will always be doing things on behalf of their humans, right? So that's why it will always be very important to understand the humans behind the agents. Maybe speaking about humans, like selling to humans. So I know you mentioned that Microsoft was a customer. I think they were kind of one of the big first customers that you have. How did you get them on board so early? Yeah, we were really lucky. we ended up hearing about this pitch competition in a niche conference around market research. And we decided to hop in and do our pitch. We ended up winning that competition.

43:05And in the audience, there were a bunch of enterprises and product was barely not working at the time. We were extremely early. It was like a couple of months in.

43:17Turner Novak:So this was like a startup pitch competition? Yeah, but for market research companies. So yeah, and you would think like, oh, if you race from Sequoia or whatever, like you're too cool to go to those pitch competitions. And a lot of founders have that mentality when we like, we showed our giant check that we won or like some of my founder friends were like, oh, why did you do that? That must've been like a waste of time. But it ended up like validating us And instead of having, you know, the typical advice for founders is to start mid-markets and then go to enterprise. Sell to startups because they'll be much faster to convert.

43:57Turner Novak:It's easier to identify the problem and like who they need to buy. Usually it's the founder, right? And they'll just like make a decision right there. Exactly. But I think that that can be a huge mistake because you can just skip that step and sell to enterprise directly. Most of the revenue is in the enterprise. And of course, it depends on what you're building. But for us, we just built it enterprise ready from day one. And we're able to start out with Microsoft, Google, P &G as one of our early customers. And a lot of very successful companies like Wiz have done that in the past because you just grow so much faster, especially in AI where the AI budgets are extremely large in enterprise, specifically traditional enterprise, that would be a piece of advice to go and build for them first.

44:50Turner Novak:So then how did you convince Microsoft? Because it's still a big company, you got to prove the use case. How did you... They were in the audience. What happened next? We had printed out this traditional survey that I was sent by IKEA. And I had it as a prop when I gave the talk and I dropped it down and you see these like pages and pages of surveys and they just felt like, wow, this is how we understand our customers. We're not treating them well enough. And they were just really bought into that idea. And then we had to sprint and build really quickly. Luckily, my co-founder is like the national champion in competitive programming in Germany.

45:35So we're able to like quickly recruit these amazing engineers. from all around the world, get them into SF and build something that worked when we were ready. The procurement process took almost a year. And so by then we actually had like a working product that was pretty good.

45:58Turner Novak:What would you say, how many total people at Microsoft did you, like different people, like individuals, did you interface with in that process? It was surprisingly simple to get the pilot done. It was just a handful of people. But now we're working with, I think, 30 teams and it's growing relatively quickly as well in the org. So it's like an infinite amount of people that can use Listen at Microsoft. So the key is land and then expand. And I'm assuming they're probably giving you feedback on the product. You probably added features based on feedback you've gotten from them, all that kind of stuff.

46:39Yeah. And that allowed us to be kind of building for other enterprises as well at the same time. It is important to have multiple enterprise customers and not just one, because then you can be kind of get stuck with them. But we always had a couple in the similar segment.

46:57Turner Novak:So you're basically telling founders, don't try to get one big enterprise customer, like try to get like three or four, like no big deal. That's easy, right? Were you able to use logos to then help you ladder up and convince other people to take you seriously because you work with this other company? Is that maybe a benefit to doing the enterprise route? Yeah, if you have Microsoft, then all the other security and compliance, those procurements, they have their own certification called SSPA. So forget about SOC 2 Type 2. You have to kind of get their own auditors to look at your stuff. It really needs to work.

47:37You can't use Delve or anything like that. And so that was a huge validation for the other enterprises that can be very slow moving. And then you use them as customer references as well.

47:51Turner Novak:Oh, yeah, that's got to be helpful. Plus, it's probably like they have a friend who works in a similar role at another company, an old coworker or something like, hey, check these guys out. Yes. So speaking of advice for other founders, I know you had a pretty interesting process for fundraising, what you recommend other founders do. What's kind of the fundraising advice that you generally give people? Yeah, less about fundraising, but more about the psychology of VCs. So one thing I found is that VCs will work much harder before they invest than after they invest. And founders should really use that to their advantage, especially in these crazy times when fundraising is a very hot market.

48:33So you should actually ask VCs to go and make a bunch of customer intros for you before they invest. And we systematized this. So we created a leaderboard that we shared in our investor updates, where you can see which VC is performing the best in terms of intros made. Not just like number of intros, but actually close one. and Ribbit ended up leading our Series B because they are like true workhorses. So a lot of their brand is not really well known, but there are these name brand VCs, they end up being a little bit complacent and they actually don't do the work that they promise that they can do.

49:22They're great at giving advice, But if you can get 10 customer enterprise intros, that can be worth a lot more. So Ribbit closed almost a million dollars in ARR for us before they led our series B. And you can actually get a large amount of pipeline from this motion. A lot of VCs will probably get annoyed by this, but it does work. They also find it kind of fun and competitive because they're very competitive in nature.

49:51Turner Novak:Yeah, hopefully they're good ones. The good ones are probably competitive. How do you actually do that in practicality though? Is it a part of the fundraise or is it like a, hey, or did you mention, hey, we think we might be raising money in three months to plant the seeds and get them in the back of their heads? Like, oh, I got to start doing some work. Or do you say like, hey, we're specifically picking our investor based on customer introductions? How do you actually tee that up in a way that lands correctly where the VCs will actually be motivated? Yeah. I mean, you have to be careful to not be too arrogant, but you can also be pretty upfront and say like, hey, you'll get a lot of VC inbound if you do a series A.

50:42And so for the... I think it only works like series A and beyond, because then it also becomes like a very significant quantum of capital. And so a lot of people will try to fight to get into your deal. But they'll reach out and then you'll say, hey, I'm not fundraising right now. But when we do, we're basically going to look at this leaderboard and we're going to pick the top folks that perform the best. And so like, would love to kind of get to work and you have to be, of course, when they do the work, you then have to show that you are building trust with them and you can't just use people, of course.

51:29But I think they also enjoy like being competitive and helping out.

51:35Turner Novak:So do you, did you like build some kind of like custom thing or is it just like a spreadsheet it's like an extension of the pipeline. We have like a vibe coded app that we share. So you raise money. I think you said you raised a hundred million total. You're obviously trying to hire people now. I'm assuming like you're trying to ramp up the team. What are you looking for in terms of like types of people, roles you're trying to fill? How do you think about adding to the team? Yeah, so hiring is one of the most competitive things in this market, especially in San Francisco. I'm not from here, so I don't have like a ton of friends.

52:13It's really been kind of a fist fight. Moved here from Sweden. And one of the ways that we have tried to differentiate in general, how I think about how you can get top tier talents if you're a small startup is by really having a distinct culture. so as i mentioned our co-founder my co-founder he is a competitive programmer so we naturally have a bunch of engineers who are like really into hard math problems and puzzles and kind of do problems on the weekends i'm old problems and we wanted to communicate that so we created this billboard that we put up in San Francisco that is just a string of random numbers.

53:03And that, if you were able to understand what that was, which, by the way, alienated most people. Like no one had no idea. Like most people had no idea what do these numbers mean.

53:16Turner Novak:Yeah, I wouldn't have known. It's literally like a URL, but it's all numbers in the URL. Like I was like, I don't know. But if you do know, it becomes this secret club. And you feel like, wow, this is very interesting. Let me go and try to understand what this is. And you realize that it was AI tokens. So you could tokenize that and you were put into this other URL where you had to act as a Berghain bouncer. So we actually had one of the problems that you do in interviews is quotas. So it's this optimization problem where you have to figure out who should be interviewed. It needs to be representative of the world.

53:58And so that's actually quite similar to being a bouncer at a club. And we kind of reframed this internal problem as a fun puzzle. We ended up going, it ended up like we spent months working on our fundraising announcement, but this ended up going much more viral than that which was unfortunate we just like took a picture with our iphone published it on x and it got millions of views we had 10 000 people actually do the puzzle and ended up now like everyone who we interview knows about this thing they don't know what our company does but they know that we did the billboard at least which i mean like that's

54:41Turner Novak:10 ,000 people that applied is like a early stage startup back at the time. Like that's, I mean, that's pretty hard to do. Yeah, no, it was, it was really cool to just see everyone trickle in. And we had people like face to compete because if you won, you were able to, we would fly out to Berlin as well to go to Berghain. It's like this like pretty legendary nightclub, I think in Berlin, like an EDM. Yeah, it's like, it's also really, it's famous for being extremely hard to get into. stay very picky about who they select. I don't think our engineer in the end, he did not actually go to Berkheim, but he did go to Berlin.

55:22Turner Novak:And so typically if you were just to not do that and you were to just say, hey, I want to like hire a recruiting agency to help me out, what do you typically pay from the recruiting agency and what do you kind of get? So I think you paid about 25 grand for this billboard. You got 10 ,000 people that did the problem and applied. If you were to go to the recruiting agency, what would you have gotten? I mean, for one engineer, you can pay$50 ,000. So it's absurdly expensive using a recruiting agency. And the big problem is that they just reach out cold with 50 other companies. So then not only do you pay the recruiting agency, but you also end up getting the most competitive candidates that are interviewing at Anthropic, OpenAI, and are getting like million dollars salaries.

56:12So with this, we're able to get like a bunch of folks that maybe the others don't know about, but they're just really excited about our culture. And that's been an advantage.

56:24Turner Novak:Yeah, that's why I think a lot of people don't always remember. Like when you see some startup that's doing some crazy thing, they're just like, oh, why did they do that? That seems kind of like a waste of time or whatever. But if you're like, I'm assuming you're not paying the same salary as like Anthropic. So you're not going to beat them by just, Hey, we'll pay more money. You have to get people that are like, Oh, this startup seems kind of interesting. Seems like a cool problem. Seems like it'd be fun to work there. I would, you know, I'll make, I'll make the jump. I'll seems like an interesting place to work.

56:56Turner Novak:Seems like a cool problem to work on. Seems like a cool product. So I feel like a lot of people, they maybe kind of glaze over that part is like, it's actually really hard to just get people to give you the time of day, even when you're trying to recruit your first 10, 50, even like sometimes first hundred, couple hundred employees, because just no one cares about you. If you're a super early stage startup, just getting started. Yeah. I mean, you should, I always start to think of it from the position of the engineer, right where they have no idea you exist there's 50 other companies growing extremely quickly and how are they going to explain it when they talk to their friends like how can you make something that you give them a cool story to explain why they joined this company specifically yeah because it's like their friends but it's also their parents because it's like okay let's say you have someone you know they went to a really prestigious school they got a job at like McKinsey or Goldman Sachs or, you know, Facebook, whatever.

57:58Turner Novak:And you're trying to convince them to kind of make this slightly crazy jump of like, Hey, you, you were like the top 1 % your whole life. And you're like, you know, you're obviously really ambitious, et cetera. And your parents are like, Hey, you know, why aren't you a doctor? Why are you doing this startup thing? Like there can be like a lot of external things that you kind of have to help them solve for too. yeah 100 um and yeah being able to make it a kind of high status and also clear why this is like a specific fit for them is that makes a huge difference and we try to hire when you think about hiring we try to find people who are kind of a little bit obsessive people who I find are great at something that could be even outside of work.

58:49They're just really passionate about it. It often translates into being successful at Les En. So we have one person, she's a race car driver. She has like eight race cars and does like drifts in Tokyo. One of our engineers built a jet engine in high school. And I also look for this almost good version of arrogance where you take a lot of pride in your work, where whatever you put out in the world, it needs to meet a certain quality bar. I find that caring about what you do is the most important, especially as the models are just getting smarter. And it actually matters less about being smart and more about having agency, being ambitious, and just caring about every single detail.

59:47Turner Novak:One interesting thread along that is one of my first... I did an internship at this big corporation in college and the CFO was just talking about what he looks for and early in your career, what do you do to stand out? And one of the things it's like, if you just spend that extra 10 minutes, like re-look at the thing you did, think of it from my perspective, do the colors look good or did you use the right font? Did you catch the last spelling error? Did Did you just spend the extra 10 or 15 or 20 minutes just like giving a shit about the thing you were about to, like the piece of work you're about to submit?

1:00:24Turner Novak:And I think about that a lot, just in everything. It's just like, okay, like I just want this to look good. Like spending an extra 10 minutes relooking at it and like maybe you redo something because you found a better way to do it. Super simple. And like, I don't know, AI won't tell you to do that. And maybe you'll think of a different, you know, lens of looking at something or framing something that wasn't there before, help someone else understand it. So. Yeah. I mean, I think a great documentary about this is called the Giro Dreams of Sushi. I don't know if you have seen that one. I actually haven't seen it, but it's, it's, he's like a guy who runs like a sushi restaurant or starts a sushi restaurant or something.

1:01:03Turner Novak:And it's like super successful. It's about this sushi chef who literally dreams of sushi and he's been doing it for 60 years and he's still obsessed with like trying to refine every single part of the detail of how you cook the rice, how you make the omelet and just has an insane quality bar. And I think with AI and you can generate AI slot now, this becomes more and more important. We see this in our interviews as well. There's a bunch of folks that will be like, oh yeah, well, I generated this case study in 10 minutes with Claude. It's good. But they actually don't look at the details. And so loving the details, that's one of our values.

1:01:47It's more important than ever.

1:01:50Turner Novak:Interesting. Well, speaking of films, I know you're really into old films. I think on your website, you have a couple of favorites that were released decades before we were both born. I think from what I saw. So what are some of your favorite movies and what do you like about them? I wanted to be a filmmaker growing up. And I think there's actually a lot of similarities with being a director, with being a startup founder, because you have this kind of interdisciplinary group that you have to align on a mission and you have to learn the technical aspects of editing as well as the creative stuff like writing great scripts, directing the actors sort of like your employees.

1:02:38You have to kind of align on your mission. So at one point, I used to watch a film a day back in high school. I think one film I really like is called Tony Erdmann. It's actually newer, but it's a German film, which is about a management consultant and her relationship with her dad. It's really funny. It has Sandra Muller in it, who was in Project Hail Mary, I think was one of her films when she became kind of famous. But I think overall, watching the classic films or reading fiction is a really good way of understanding what people want. And storytelling is a really important skill if you are a startup founder.

1:03:36So I recommend everyone to watch Digmar Bergman. It's a Swedish film director.

1:03:42Turner Novak:Interesting. A movie about a management consultant. Okay, well, I'll throw a link in the description for people to find it. So a different topic. But I remember hearing that you guys have a certain harness that you made for the agents at Listen. So what exactly is that? Yeah, so an agent harness is kind of the framework that the agent can use to do tool calling and the knowledge management. and what we found was that every other harness is built around a file system. So Cloud Code, for example, will use Cloud MD and that's how it kind of has memory and figures that out. For us, that's the wrong architecture, specifically for statistical analysis.

1:04:34So we, because we think that the right way of building a harness is a table because you can kind of operate on it as a pandas data frame which is a a tool in python so you're able to kind of every row is a response and then every column is a is a feature so every row is like an interview and then you can extract information for every single interview so you can tell our agent to kind of if you want to quantify something it can run a sub-agent for every single response and classify like, does this person like my product or not? Even if you collected like open-ended interviews, if that makes sense.

1:05:18And then you can easily do like aggregated stats. You can run correlations between columns. That's like much harder in a file system. So that's one thing that I've seen that these vertical AI companies can do is essentially look at the job that you're trying to do as an agent and really perfect the hardness, perfect the workflow around that job. And you can get much more juice out of the models than the vanilla model companies.

1:05:53Turner Novak:Interesting. So it would basically be like if I'm Doritos and I'm asking some questions about like a new flavor, existing flavors, how I feel about the, like, I don't, you don't specifically ask me, Turner, do you like Doritos? But you will be able to tell if I do like Doritos based on how I answered other questions, essentially. Yeah. It's all open-ended and you feed all of that into the lemon and it's able to predict how you like or not like something. I actually, I wanted to ask you something. So I know you do this fellowship where you bring, and talking about yourself, I know you're from Sweden, moved to the U S you actually run this fellowship program for other Swedes, helping them move to San Francisco.

1:06:32Turner Novak:What's the program and what do you guys do? Yeah, I run this program called Velocity Fellows. And I always struggled being kind of the only one obsessed with startups back in Stockholm and wanted to create a space where people like that can find other like-minded founders and then bring them to SF to kind of scale their ambition. So the goal is not for them to move to SF, because I don't want to like, increase brain drain, but hopefully to bring the SF spirit back to Sweden. So we had like Max Unistrand, who's now the founder of Legora, used to be a kind of intern at my company as well. He was part of the batch one.

1:07:19And we have a lot of them have now raised money and we connect them with like, there's a bunch of Swedish folks in Silicon Valley as well, like Ali Gusti, who's the CEO and founder of Databricks. He's Swedish. Eric Bernards, who's at Modal, he's Swedish as well. And it's cool. We're seeing like resurgence of the Swedes.

1:07:43Turner Novak:Nice. I was going to say, yeah, I had Eric on the podcast a couple months ago. He's really fun. Talking about like clips and stuff, he actually, he had one of the most viral clips of the podcast. It was about CO2 levels in the office, the most random topic, but it got thousands of likes on Twitter, like a couple hundred thousand views. I think it was close to a billion views. It was like people and people were like chiming in like, yes, CO2 levels, like you need to manage the CO2 level in your office. Like it actually has a huge impact on your work productivity. And I was like, wow, did not know that this was such a big deal.

1:08:16Turner Novak:But it's true, I guess. His he's like very big on like they have CO2 monitors in the office and make sure that CO2 levels don't get too high because it impacts your brain and makes you less productive. So it's like, huh, all right. Interesting. One other fun fact that I remember hearing about you is I think your brother is the founder of SoundCloud. That's true. He's 16 years older than me. Okay. Yeah. I used to be a pretty heavy SoundCloud user, just a lot of like EDM remixes and stuff. Less so now. They're just like less people post on SoundCloud, I feel like. But I definitely have fond memories of my first job.

1:08:56Turner Novak:I was an analyst at a bank, just listening to Chain Smokers remixes and Avicii remixes on SoundCloud for 10 hours a day. No, yeah, SoundCloud is obviously a big part of my childhood being like seeing in building that company, the things to do, the things not to do. got to visit the office when I was very young. And I'm also competitive. And so I want to try to build something that's bigger than my brother's company. But he's moved now to the Bay Area as well. So we spent a lot of time together. Oh, cool. And I think one other thing I heard you say, you've mentioned before that you guys have a no shoes policy in the office.

1:09:44Turner Novak:That's a pretty, There's a lot of... You go on the internet, people have strong opinions of shoes versus no shoes. So what's the shoe policy? Yeah, so having no shoes makes it much more comfortable. It feels like you're at home. And that allows for more of an academic environment, I think, which is one of our values. like that folks be able to like have free discussions and you can sit in the sofas and be more open. We also have listen branded slippers. So if you do need some shoes, when you come in, we help you swap from your sneakers to your, to our slippers. And, but it seems to be a very controversial topic, which I don't fully understand why it's obviously much better to not have shoes in the office.

1:10:38Yeah.

1:10:38Turner Novak:Well, I think the thing that I think is kind of crazy is if you walk through San Francisco, you know, not the cleanest city in the world, and then you go into an office, like you are walking the same shoes that were on the ground that people are partaking in the external outside activities that happen on the streets in San Francisco that you then do in an office. Like I can see the value behind it. yeah and it's from my high school we didn't have any shoes on there as well this is like hippie high school in sweden yeah whoa we also only had vegetarian food um so it was um a school that was controlled by the students so if the students voted for something in the majority it would happen they had to stop that after a while because students ended up abolishing homework and things like that.

1:11:34Turner Novak:But was that the, what was the craziest thing that happened? Was it the no homework? I think that's when they had to pull it back. Um, but the vegetarian food was, was also a big one. Like, and it was amazing. It was so delicious. Um, but that's, that's the inspiration behind no shoes. Interesting. Okay. Where, uh, where can people find you? I think you post on Twitter? Are you pretty active on LinkedIn? Yeah, you can follow me on X and on LinkedIn, Alfred Wallforce, or go to ListenLabs.ai and sign up for a demo. And we're also hiring for engineers, salespeople. We're around 60 people and want to be 150 by the end of the year.

1:12:18So trying to scale very quickly.

1:12:21Turner Novak:Nice. Well, we'll throw links, all those in the description and people can find you. This is a lot of fun. Thanks for doing it. Yeah, thank you so much.

1:12:57Turner Novak:with Hans, who started Secondary's Firm Industry Ventures, which was just acquired by Goldman Sachs, with Dan at Gutter Capital, and a conversation I recorded at Allocates Beyond Summit featuring observations from 15 GPs and LPs on what they're seeing on the ground of the venture capital market today. If you don't want to miss any of these, subscribe to my newsletter, The Split, linked in the description to get each episode plus a transcript emailed directly to your inbox every week. Thanks again for listening. See you next time.

1:13:27Bye.