Descript Transcript
[00:01:52] AI Charlie: Welcome back. This is Charlie, your AI cohost. As AI engineers, we have a special focus on coding agents, fine tuning, and synthetic data. And this week, it all comes together with the launch of Cosign's Genie, which reached 50 percent on SWE Bench Lite, 30 percent on the full SWE Bench, and 44 percent on OpenAI's new SWE Bench Verified.
[00:02:17] All state of the art results by the widest ever margin recorded compared to former leaders Amazon Q and US Autocode Rover. And Factory Code Droid. As a reminder, Cognition Devon went viral with a 14 percent score just five months ago. Cosign did this by working closely with OpenAI to fine tune GPT 4. 0, now generally available to you and me, on billions of tokens of code, much of which was synthetically generated.
[00:02:47] Alistair Pullen: Hi, I'm Ali. Co founder and CEO of Cosign, a human reasoning lab. And I'd like to show you Genie, our state of the art, fully autonomous software engineering colleague. Genie has the highest score on SWBench in the world. And the way we achieved this was by taking a completely different approach. We believe that if you want a model to behave like a software engineer, it has to be shown how a human software engineer works.
[00:03:15] We've designed new techniques to derive human reasoning from real examples of software engineers doing their jobs. Our data represents perfect information lineage, incremental knowledge discovery, and step by step decision making. Representing everything a human engineer does logically. By actually training Genie on this unique dataset, rather than simply prompting base models, which is what everyone else is doing, we've seen that we're no longer simply generating random code until some works.
[00:03:46] It's tackling problems like
[00:03:48] AI Charlie: a human. Alistair Pullen is CEO and co founder of Kozen, and we managed to snag him on a brief trip stateside for a special conversation on building the world's current number one coding agent. Watch out and take care.
[00:04:07] Alessio: Hey everyone, welcome to the Latent Space Podcast. This is Alessio, partner and CTO of Resonance at Decibel Partners, and I'm joined by my co host Swyx, founder of Small. ai.
[00:04:16] swyx: Hey, and today we're back in the studio. In person, after about three to four months in visa jail and travels and all other fun stuff that we talked about in the previous episode.
[00:04:27] But today we have a special guest, Ali Pullen from Cosign. Welcome. Hi, thanks for having me. We're very lucky to have you because you're on a two day trip to San Francisco. Yeah, I wouldn't recommend it. I would not
[00:04:38] Alistair Pullen: recommend it. Don't fly from London to San Francisco for two days.
[00:04:40] swyx: And you launched Genie on a plane.
[00:04:42] On plain Wi Fi, um, claiming state of the art in SuiteBench, which we're all going to talk about. I'm excited to dive into your whole journey, because it has been a journey. I've been lucky to be a small angel in part of that journey. And it's exciting to see that you're launching to such acclaim and, you know, such results.
[00:05:01] Alistair and Cosine intro
[00:05:01] swyx: Um, so I'll go over your brief background, and then you can sort of fill in the blanks on what else people should know about you. You did your bachelor's in computer science at Exeter.
[00:05:10] Speaker 6: Yep.
[00:05:10] swyx: And then you worked at a startup that got acquired into GoPuff and round about 2022, you started working on a stealth startup that became a YC startup.
[00:05:19] What's that? Yeah. So
[00:05:21] Alistair Pullen: basically when I left university, I, I met my now co founder, Sam. At the time we were both mobile devs. He was an Android developer. iOS developer. And whilst at university, we built this sort of small consultancy, sort of, we'd um, be approached to build projects for people and we would just take them up and start with, they were student projects.
[00:05:41] They weren't, they weren't anything crazy or anything big. We started with those and over time we started doing larger and larger projects, more interesting things. And then actually, when we left university, we just kept doing that. We didn't really get jobs, traditional jobs. It was also like in the middle of COVID, middle of lockdown.
[00:05:57] So we were like, this is a pretty good gig. We'll just keep like writing code in our bedrooms. And yeah, that's it. We did that for a while. And then a friend of ours that we went to Exeter with started a YC startup during COVID. And it was one of these fast grocery delivery companies. At the time I was living in the deepest, darkest countryside in England, where fast grocery companies are still not a thing.
[00:06:20] So he, he sort of pitched me this idea and was like, listen, like I need an iOS dev, do you fancy coming along? And I thought, absolutely. It was a chance to get out of my parents house, chance to move to London, you know, do interesting things. And at the time, truthfully, I had no idea what YC was. I had no idea.
[00:06:34] I wasn't in the startup space. I knew I liked coding and building apps and stuff, but I'd never, never really done anything in that area. So I said, yes, absolutely. I moved to London just sort of as COVID was ending and yeah, worked at what was fancy for about a year and a half. Then we brought Sam along as well.
[00:06:52] So we, Sam and I, were the two engineers at Fancy for basically its entire life, and we built literally everything. So like the, the front, the client mobile apps, the, the backends, the internal like stock management system, the driver routing, algorithms, all those things. Literally like everything. It was my first.
[00:07:12] You know, both of us were super inexperienced. We didn't have, like, proper engineering experience. There were definitely decisions we'd do differently now. We'd definitely buy a lot of stuff off the shelf, stuff like that. But it was the initial dip of the toe into, like, the world of startups, and we were both, like, hooked immediately.
[00:07:26] We were like, this is so cool. This sounds so much better than all our friends who were, like, consultants and doing, like, normal jobs, right? We did that, and it ran its course, and after, I want to say, 18 months or so, GoPuff came and acquired us. And there was obviously a transitionary period, an integration period, like with all acquisitions, and we did that, and as soon as we'd vested what we wanted to vest, and as soon as we thought, okay, this chapter is sort of done, uh, in about 2022, We left and we knew that we wanted to go alone and try something like we'd had this taste.
[00:07:54] Now we knew we'd seen how a like a YC startup was managed like up close and we knew that we wanted to do something similar ourselves. We had no idea what it was at the time. We just knew we wanted to do something. So we, we tried a small, um, some small
