Y Combinator Startup Podcast
00:27:09 · Harj Taggar, Paul Copplestone
[00:27:09] S2: I guess, do you think about the personal agent world in that... Through that perspective then if sort of coding agents want Supabase, that's very clear, do you think about the next phase as being assuming we all have personal databases, like do you need to do anything to make sure that it's easy for personal agents to talk to you? Is that any different, or is it based the same thing? [00:27:29] S0: Yeah. I'm very bullish now on personal agents after a week of my EA being away and me learning exactly what they can do. And now I'm just kind of obsessed with the idea of building for personal agents. And it's true that I I said to our company, like, we're code first. Like, we're building for humans that are doing code first, but, actually, we're doing it for agents. Those humans pretty soon won't actually be interacting. So, like, the way I've been building is I'm using dots. I just sit on a phone call with it all the time and I'm chatting to it and saying, oh, can you make these changes? And my my kind of heuristic to make sure I'm building or dogfooding correctly is that I just don't look at the code. And I don't allow myself ever to check the architecture except when I wanna check that it's actually building things correctly. But when I'm making these changes, I just wanna talk to it and see what it can change. [00:28:25] S2: What of us?
00:28:25 · Harj Taggar, Paul Copplestone, S3, Garry Tan
[00:28:25] S2: What of us? [00:28:26] S0: What of It's crazy as well. Like, it's doing some phenomenal stuff that I just... It it feels like something out of a sci fi movie from, like, two years ago. But now, like, I'm having so many of these moments that I'm just mind blowing. If you had shown someone this six months ago, they would've just thought it was impossible. [00:28:48] S3: Yeah. You... I was at your Supabase Select conference a couple weeks ago, which is an awesome conference, by the way. And I had this, like, kinda crazy thought. I was sitting there, and you're talking about all these awesome new features that Supabase was putting out. And I was like, you know, if I was building an app on Supabase, maybe? I wouldn't even care to understand all these awesome features myself. I wanna, like, send my agent to Supabase Select to, like, listen to all the awesome new features and figure out how to integrate them into my code base. [00:29:17] S0: Oh, that's wild. A truly virtual conference. We should do that. A conference just for agents. [00:29:22] S3: For agents. Yeah. A conference for agents. [00:29:24] S0: They all turn up live. [00:29:26] S3: Live. They can like talk to each other, exchange tips on like, you [00:29:29] S0: know, this book live. [00:29:30] S2: Wow. We can bridge it. You can sort of airdrop in like the markdown files for everyone. But you have to be there in, like, the the audience to get it. [00:29:38] S1: We'll call it the artifactory. Yeah. [00:29:42] S0: Oh, man. I quite like that.
00:29:42 · Paul Copplestone, Garry Tan
[00:29:42] S0: Oh, man. I quite like that. [00:29:43] S1: We talked about before this, like, there's just a bunch of things that you've been doing that, like, really make agents like you better, like error messages. You know. You know, what was that key insight? And, like, are there a bunch of other things that you're doing that kind of obvious on the one hand, on the other hand, like, nobody's doing yet? [00:30:01] S0: Yeah. So, like, the error messages... Yeah. I was at a DevToom Center last night, I was surprised that this wasn't so obvious. But we really dug into our error messages so that you can take a use case like, oh, I want to build an application. And generally, like, developer tomb error messages of like 500 something basic. And you just know there's a problem, but now you actually want the agents to, like, remedy themselves. So inside the error messages, you are literally telling them how to fix this error or to go and read some docs to fix the error. So a lot of it is just picking your use case and just trying to do these loops on improving your own error messages so that you can one shot basically anything that you want. [00:30:45] S1: Yeah. That makes sense. Basically, you know, if you're being used by an agent, the agent will encounter a error message. In the past, a human would have to go Google that and do a bunch of stuff. But here, you can just say, oh, like, you're just, you know, out of inode, so you should probably go do this.
00:30:45 · Garry Tan, Paul Copplestone, S3
[00:30:45] S1: Yeah. That makes sense. Basically, you know, if you're being used by an agent, the agent will encounter a error message. In the past, a human would have to go Google that and do a bunch of stuff. But here, you can just say, oh, like, you're just, you know, out of inode, so you should probably go do this. [00:31:02] S0: Yeah. Because there's, like, two stages to winning the agents. The first is being picked by the model. And then the second is, like, if they get stuck, you don't want them to switch away. So guaranteed they'll get stuck at some point. You just want to give them the tools to to remedy. So it's not actually that complicated. They're smart enough now that they can figure out a lot of stuff, but you do have to reengineer a lot of what previously was just for humans and developers for agents. [00:31:31] S3: Do you test your project... Your product for agents? Like, do you have agents, like, try to use Supabase and then, like, report all the places where they had trouble?
00:31:31 · S3, Paul Copplestone
[00:31:31] S3: Do you test your project... Your product for agents? Like, do you have agents, like, try to use Supabase and then, like, report all the places where they had trouble? [00:31:39] S0: Yeah. Yeah. Yeah. All the different models. We got an eval's website where we see which models are building the best and got our own eval suite internally. We eval as well which ones are best just with Postgres, so that's interesting for the industry and build skills around it so that they can adopt them and actually develop with Postgres. So, yeah, it... It's a lot of testing. And in fact, pretty much now a lot of our engineering effort is going into, like, testing specific use cases rather than the building. We wanna think through how people are building and build suites around that. [00:32:16] S3: Garry, when you built g Brain, you built it on top of Supabase. Can talk about, like, how did you decide to build it on top of Supabase? Well, how was it like for you to be a user?
00:32:25 · Garry Tan
[00:32:25] S1: Yeah. I guess, I mean, g brain was like a distillation of a bunch of stuff I wrote. Like, I first worked on Gary's List, which is like agentic. It was like an agentic newsroom. And I had to build up, like, basically a research system for rag and for like policy and papers, and it would like read every newspaper article about a given topic, and then try to synthesize. And then I took that and then distilled it into open source retrieval for OpenClaw. Because the idea was like, okay, I had it in Rails, and that was kinda like a Foxconn factory for agents. It's like, I treated every LLM call as very expensive and incredibly dangerous. And then when I saw OpenClaw, it was this, you know, it was kinda like Esselen. Was just like a very different way for, you know, you to work with AI, because then it's like, oh, the AI's smart. It's not trying to hurt you, and then you should just let it, like, do stuff, and you should just give it data and capabilities. And so I recognized a million tokens is actually an incredible working memory. Like, most human beings only have, like, seven plus or minus two things that they can keep in their head, whereas, like, a million tokens is, like, basically, like, three Harry Potter books. But then even that is, like, a little bit limiting for the agents. Like, keeping the correct three Harry Potter books in your context at any given time is a real challenge. So it felt like there was this thing in between. For a company brain or, like, my personal stuff, it's like, I have half a million Markdown files on, like, every communication I've ever had. And so this thing really does have context on me and everything that I care about. But then at any given moment, like, the agent, they might need like two pages or something. It's it's a retrieval problem. And so at that point, you know, at many, I think it's like 30, 40 gigabytes, You need p g vector, you need Postgres, you know, you can do graph databases very easily in SQL. And so that's what g brain is. It's like, you know, rag retrieval plus like a a graph database that creates, like structure out of unstructured data. And so it's like dynamic schemas. Like a little bit like what I remember Palantir was back in the day. Yeah.