Ep 60: Swyx and Alessio (Latent Space) on What has PMF Today, Google is Cooking & GPT Wrappers are Winning

To unpack some of the most topical questions in AI, I’m joined by two fellow AI podcasters: Swyx and Alessio Fanelli, co-hosts of the Latent Space podcast. We’ve been wanting to do a cross-over episode for a while and finally made it happen. Swyx brings deep experience from his time at AWS, Temporal, and Airbyte, and is now focused on AI agents and dev tools. Alessio is an investor at Decibel, where he’s been backing early technical teams pushing the boundaries of infrastructure and applied AI. Together they run Latent Space, a technical newsletter and podcast by and for AI engineers.

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Jacob Effron?

  Today, my partner Jordan and I have a special episode of unsupervised learning, a crossover with one of our favorite AI podcasts, Late in Space. If you're not already a listener, Late in Space is a technical newsletter and podcasts by and for AI engineers. It had over 2 million downloads in 2024, and it's become a go-to resource for anyone who wants to understand the cutting edge of AI infrastructure, tooling, and product. If you like this show, it's definitely worth checking out. Given we've all spent a lot of time talking to some of the sharpest minds in AI, we thought it'd be fun to interview each other. In this episode, we dig into the questions we're constantly thinking about. What surprised us most last year? What we're paying most attention to right now? How we think about defensibility at the app layer? And which public companies were long or short on? It's a different kind of episode, and I think you'll really enjoy it. Now, here's my conversation with Swyx and Alessio from Late in Space.

• 0:45Jacob Effron?

  Well, thanks so much for doing this, guys. I feel like we've been excited to do a collab for a while.

• 0:49Speaker 01

  I love crossovers. Yeah, this is great. Like, the ultimate meta about just podcasters talking to other podcasters.

• 0:55Jacob Effron?

  Yeah, it's a lot of podcasts all the way up. I figured we'd have a pretty free-ranging conversation today, but brought a few conversation starters to kick us off. And so I figured one interesting place to start is, you know, obviously it feels like this world is changing like every few months. Wondering if you guys reflect on the past year, like, what surprised you the most?

• 1:12Speaker 02

  I think definitely recently models. We're kind of on the right here. We're like calling... Oh, that. Well, I think there's like the what surprised us in a good way, maybe in a bad way, I would say. In a good way, recently models, and I think the release of them right after the new reps scaling is that. Talked about Ilya. I think there was maybe like a little it's so over and then we're so back and like such a short period of time.

• 1:36Jacob Effron?

  It was really fortuitous timing, like right as pre-training died. And obviously, I'm sure within the labs, they knew pre-training was dying and had to find something. But, you know, from the outside, it was, it felt like one right into the other.

• 1:47Speaker 02

  Yeah, yeah, exactly. So that was a good surprise.

• 1:50Speaker 01

  I would say if you want to make that comment about timing, I think it's suspiciously neat that like, we know that Strawberry was being worked on for like two years-ish. Like, we know exactly when Gnome joined OpenAI, and that was obviously a big strategic bet by OpenAI. So like, for it to transition so nicely when like pre-training is kind of tapped out to into like, oh, now inference time is the new scaling law, is like very convenient. I like, if there were an Illuminati, this would be what they planned.

• 2:20Jacob Effron?

  Or a simulation or something.

• 2:23Speaker 02

  Yeah.

• 2:24Jacob Effron?

  Then you said open source as well?

• 2:25Speaker 02

  Yeah, well, no, I think like open source, yeah, we're discussing this on the negative. I would say the relevance of open source.

• 2:31Speaker 01

  Specifically open models.

• 2:33Speaker 02

  Yeah, I was surprised by the lack, by the lack of adoption. I mean, people use it, obviously, but I would say nobody's really like a huge fanboy. You know, I think the local Lama community and some of the more obvious use cases really like it. But when we talk to like enterprise folks, it's like, it's cool, you know? And I think people love to argue about licenses and all of that. But the reality is that it doesn't really change the adoption path of AI, so.

• 2:59Speaker 01

  Yeah, the specific stat that I got from Anker from Brain Trust in one of the episodes that we did was, I think he estimated that open source model usage in work in enterprises is at like 5% and going down.

• 3:11Jacob Effron?

  And it feels like you're basically, all these enterprises are in like use case discovery mode where it's like, let's just take what we think is the most powerful model and figure out if we can find anything that works. And, you know, so much of it feels like discovery of that. And then right as you've discovered something, a new generation of models are out and so you have to go do discovery with those. And, you know, I think obviously we're probably optimistic that the, that the open source models increase in uptake. It's funny, I was going to say my biggest surprise in the last year was open source related, but it was just how fast open source caught up on the recent models. It was kind of unclear to me, like over time, whether there would be, you know, a compounding advantage for some of the closed source models. Where in the, okay, in the early days of scaling, you know, there was a tight time loop. But over time, you know, would, would the gap increase? And if anything, it feels like a trunk, you know?

• 3:53Jacob Effron?

  And I think DeepSeek specifically was just really surprising in how, you know, in many ways, if the value of these model companies is like, you have a model for a period of time and you're the only one that can build products on top of that model while you have it. Like, God, that time period is a lot shorter than, than I thought it was going to be a year ago.

• 4:08Speaker 01

  Yeah, I mean, again, I don't like this label of how fast open source caught up because it's really how fast DeepSeek caught up, right? And now we have like, I think some evidence that DeepSeek is basically going to stop open sourcing models. So like, there's no team open source, there's just different companies and they choose to open source or not. And we got lucky with DeepSeek releasing something and then everyone else is basically distilling from DeepSeek. And those are, distillations catching up is such an easier, lower bar than like actually catching up, which is like. You like from scratch, you're training something that like is competitive. On that front, I don't know if that's happening. Like basically, the only player right now is we're waiting for Llama 4.

• 4:47Speaker 03

  I mean, it's always in order of magnitude cheaper to replicate what's already been done than to create something fundamentally new. And so that's why I think DeepSeek overall was overhyped, right? I mean, obviously, it's a good open source new entrant, but at the same time, there's nothing new fundamentally there other than sort of doing it, executing what's already been done really well.

• 5:05Speaker 01

  Yeah.

• 5:05Speaker 03

  Right? So.

• 5:06Speaker 02

  Well, but I think the traces is like maybe the biggest thing. I think most previous open models is like the same model, just a little worse and cheaper.

• 5:15Speaker 01

  Yeah.

• 5:15Speaker 02

  Like R1 is like the first model that had the full traces. So I think that's like a nat, unique thing in open source. But yeah, I think like we talked about DeepSeek in the, our end of year 2023 recap. And we're mostly focused on cheaper inference. Like, we didn't really have.

• 5:31Speaker 01

  DeepSeek V3 was out then and we were like, there was already like talking about fine green, uh, mixture of experts and all that.

• 5:36Speaker 03

  That's a great receipt to have to be like, yeah, end of year 23.

• 5:39Speaker 02

  We were there.

• 5:39Speaker 03

  Yeah.

• 5:39Jacob Effron?

  That's a, that's a, that's, that's an impressive one.

• 5:41Speaker 01

  You follow the right whale believers on Twitter. It's, it's like pretty obvious. I actually had like, uh, so you know, I used to be in finance and, and a lot, a lot of my hedge fund and PE friends called me up. They were like, why didn't you tip us off on DeepSeek? And I'm like, well, I mean, it's been there. It's actually like kind of surprising that like Nvidia like fell like what 15% in one day.

• 6:01Jacob Effron?

  Yeah.

• 6:02Speaker 01

  Uh, because of DeepSeek. And I, I think it's just like whatever the market, public market narrative decides is a story becomes the story. But really like the technical movements are usually one to two years in the making before that.

• 6:13Speaker 03

  Basically, these people were telling on themselves and they didn't listen to your podcast. They've been on the end of year 23.

• 6:18Speaker 01

  No, no, no, like, uh, yeah, we weren't, we weren't like banging the drum. So like, it's also on us to be like, no, like this, this is an actual tipping point. And I think I like as people who are like, our function as podcasters and industry analysts is to. Raise the bar or focus attention on things that you think matter. And sometimes we're too passive about it. And I think I was too passive there. I'd be happy to own up on that.

• 6:38Jacob Effron?

  No, I feel like over time, you guys have moved into this marching role of like taking stances of things that are or aren't important. And, you know, I feel like you've done that with MCP, Blade and a bunch of things.

• 6:48Speaker 01

  Yeah, so the general push is AI engineering, you know, like gotta wrap the shirt. And MCP is part of that. But like the general movement is what can engineers do above the model layer to augment model capabilities? And it turns out it's a lot. And it turns out we went from like making fun of GPT rappers to now, I think the overwhelming consensus, GPT rappers is the only thing that's interesting.

• 7:09Jacob Effron?

  Yeah. I remember like Arvin from Perplexity came on our podcast and he was like, I'm proudly a rapper. Like, you know, it's like anyone that's like talking about like, you know, differentiation, like pre-product market fit is like a ridiculous thing to say. Like build something people want and then over time you can kind of worry about that.

• 7:24Speaker 01

  Yeah, I interviewed him in 2023 and I think he may have been the first person on our podcast to like proudly be a GPT rapper. And yeah, obviously he's built a huge business on that.

• 7:33Jacob Effron?

  Totally. Now we all can't get enough of it.

• 7:35Speaker 01

  I have another one for, so that was the last year's one. And we perhaps individual answers just to be interesting.

• 7:40Jacob Effron?

  In the same Uber on the way up?

• 7:42Speaker 01

  Yeah.

• 7:42Jacob Effron?

  You're just like in the.

• 7:43Speaker 01

  I was driving too. So it was like actually. I mean, it was a Tesla, so I mostly drove it. Mine was actually, it's interesting that low-code builders did not capture the AI builder market. Right? AI builders being Boll and Lovable, low-code builders being Zapier, Airtable, Retool, Notion, any of those like when you're not technical, you can build software.

• 8:04Jacob Effron?

  Yeah.

• 8:05Speaker 01

  Somehow not all of them missed it. Why?

• 8:08Jacob Effron?

  Yeah.

• 8:09Speaker 01

  It's bizarre. Like they should have the DNA. I don't know. They should have, they already have the reach. They already have the distribution. Like why? I have no idea.

• 8:17Jacob Effron?

  The ability to fast follow too. Like I'm surprised.

• 8:19Speaker 01

  Yeah, there's just nothing.

• 8:21Jacob Effron?

  Yeah. What do you make of that?

• 8:23Speaker 01

  I, it seems that, and you know, not to come back to the AI engineering picture, like it takes a certain kind of

• 8:30Speaker 01

  founder mindset or AI engineer mindset to be like, we will build this from whole cloth and not be tied to existing paradigms, I think. Cause like if I was, if I'm to, you know, you know, Wade or who's the Zapier person that you know? Mike?

• 8:45Jacob Effron?

  Mike.

• 8:46Speaker 01

  Who has left Zapier?

• 8:47Jacob Effron?

  The Zapier person.

• 8:50Speaker 01

  Yeah, like, you know, Zapier, when they decided to do Zapier AI, they were like, oh, you can use natural language to make Zapier actions, right? When Notion decided to do Notion AI, they were like, oh, you can like, you know, write documents or, you know, fill in tables with AI. Like, they didn't do the next step because they already had their base. And they were like, let's improve our baseline. And the other people who actually tried to create a phone call were like, we got no prior preconceptions, like, let's see what we can, what kind of software people can build with, like, from scratch, basically. I don't know. That's my explanation. I don't know if you guys have any retros on the AI builders.

• 9:27Jacob Effron?

  Yeah, or did they kind of get lucky getting, you know, starting that product journey like right as the models were reaching the inflection point?

• 9:35Speaker 01

  There's the timing issue. Yeah, yeah, yeah, yeah. Yeah, I don't know. Like, to some extent, I think the only reason you and I are talking about it is that they, both of them have reported like ridiculous numbers, like zero to 20 million in three months, basically, both of them.

• 9:49Jacob Effron?

  Jordan, did you have a big surprise?

• 9:51Speaker 03

  Yeah, I mean, some of what's already been discussed. I guess the only other thing would be on the Apple side in particular. I think, I think, you know, for the last-

• 9:59Jacob Effron?

  Those text message summaries, like, phew.

• 10:01Speaker 01

  But they're funny.

• 10:02Jacob Effron?

  They're funny and how back they are and how off they are. They're viral.

• 10:05Speaker 03

  Yeah, I mean, so like, for the last couple of years, we've seen so many companies that are trying to do personal assistants, like all these various consumer things. And one of the things we've always asked is, well, Apple is in prime position to do all this. And then with Apple Intelligence, they just totally messed up in so many different ways. And then the whole BBC thing, saying that the guy shot himself when he didn't. And just like, there's just so many things at this point that I would have thought that they would have ironed up their AI products better, but just didn't really catch on.

• 10:33Jacob Effron?

  You know, second on this list of generally overly broad opening questions would be, anything that you guys think is kind of like overhyped or underhyped in the AI world right now?

• 10:42Speaker 02

  Overhyped agents framework.

• 10:44Speaker 01

  Not naming any particular ones.

• 10:46Speaker 02

  I'm sorry. Not, not, yeah, exactly. It's not, I would say they're just overall a chase to try and be the framework when the workloads are like in such flux.

• 10:57Speaker 01

  Yeah.

• 10:57Speaker 02

  That I just think is like so hard to reconcile the two. I think what Harrison and Langchain has done so amazingly is like uh product velocity, like, you know, uh, the initial obstructions were maybe not the ending obstruction, but like they were just releasing stuff every day trying to be on top of it. But I think now we're like past that. Like what people are looking for now is like something that they can actually build on and stay on for the next couple of years. And we talked about this with Brett Taylor on our episode and it feels like it's like the jQuery era.

• 11:28Speaker 01

  Yeah.

• 11:28Speaker 02

  Of like agents and L L M's. It's like, it's kind of like, you know, single file, big frameworks, kind of like a lot of helpers, but maybe we need React. And I think people are just trying to build still jQuery. Like, I don't really see a lot of people doing React like.

• 11:42Speaker 01

  Yeah. Um, maybe the only modification I made about that is maybe it's too early even for frameworks at all.

• 11:49Jacob Effron?

  Yeah, I think there's enough stability uh in the underlying model layer and patterns to have this.

• 11:54Speaker 01

  The thing is the protocol and not the framework.

• 11:56Jacob Effron?

  Yeah.

• 11:56Speaker 01

  Uh, because frameworks inherently embed protocols, but if you just focus on a protocol, maybe that works. And obviously MCP is the current leading uh area. And you know, I think the comparison there would be instead of just jQuery, it is XML HTTP requests, which is like the thing that enabled Ajax. Um, and that was the, the, the, the sort of inciting incident for JavaScripts uh being popular as a language.

• 12:18Speaker 03

  I would largely agree with that. I mean, I think on the, the React side of things, I think we're starting to see more frameworks sort of go after more of that. I guess like MASTRA is sort of like on the TypeScript side and more of like a sort of.

• 12:29Speaker 01

  MASTRA, yeah.

• 12:30Speaker 03

  Yeah, yeah. The traction is really impressive there. And so I think um we're starting to see more surface there, but I think there's still a big opportunity.

• 12:37Jacob Effron?

  What do you have for a, for an over or underhyped?

• 12:40Speaker 03

  On the underhyped side, um, you know, actually I, I know I mentioned Apple already, but I think the private cloud compute side with PCC, I actually think that could be really big. It's under the radar right now, but in terms of basically bringing. The on-device sort of security to the cloud. They've done a lot of architecturally interesting things there. Who's they? Apple. Oh, okay. On the PCC side. And so I actually think that. See a negative on Apple Intelligence, but positive on Apple Cloud. On the more of the local device, sort of. I think there will be a lot of workloads still on-device. But when you need to speak to the cloud for larger LLMs, I think that Apple has done really interesting thing on the privacy side.

• 13:18Speaker 02

  We did the seat of a company that does that.

• 13:20Speaker 03

  Yeah, especially as things become more consumerized. Yeah, no, I was like, let's go, Jordan. That's cloning before this episode. Tell me about that company after. We'll chat after.

• 13:33Speaker 02

  But, but yes, I, I think that's like the unique. The thing about LLM workloads, it's like you just cannot have everything be single-tenant. Because you just cannot get enough GPUs. Like even, like, large enterprises are used to having VPCs and like everything runs privately. But now you just cannot get enough GPUs to run in a VPC. So I think you're gonna need to be in a multi-tenant architecture. And you need, like you said, like, single-tenant guarantees in multi-tenant environments. So, um, yeah, it's an interesting space.

• 14:00Speaker 03

  Yeah, what about you, Sikes?

• 14:02Speaker 01

  Under hyped, I want to say memory, just like stateful AI.

• 14:10Speaker 01

  As part of my keynote on, on, it's like every, every conference I do, I do a keynote. And I tried to do the task of like define an agent. You know, always.

• 14:21Speaker 03

  Evergreen content for a keynote.

• 14:24Speaker 01

  But I did it in a, in a way that was like, um, I think like what a researcher would do. Like you, you survey what people say and then you sort of categorize and, and go like, okay, this is the, the, uh, what everyone calls agents. And here are the groups of definitions pick and choose. Right? And then it was very interesting that the week after that, OpenAI launched their Agent SDK. And kind of formalized what they think agents are. Qualify also did the same with us. And. None of them have memory.

• 14:53Jacob Effron?

  Yeah.

• 14:54Speaker 01

  Very strange. The, pretty much like the only big lab, obviously there's conversation memory, but there's not memory, memory like in like a, like a let's store a knowledge graph of facts about you and like, you know, exceed the, the context length.

• 15:10Speaker 01

  And here's the, if you, if you're, if you look closely enough, there's a really good implementation of memory inside of MCP. When they launched with the initial set of servers, they had a memory server in there, which I, I would recommend as like, that's where you start with memory. But I think like, if there was a better memory abstraction, then a lot of our agents would be smarter and could learn on, on the job, which is something that we all want. And for some reason, we've all just like ignored that because it's just convenient to and.

• 15:41Jacob Effron?

  But do you think it's being ignored or it's just a really hard problem? And like lots of, and I feel like lots of people are working on it. Just feels like it's, it's proven more challenging.

• 15:48Speaker 01

  Yeah, yeah, yeah. So, so Harrison has Langmem, which I think now he's like, you know, relaunched again. And then we had Letter come speak at our, our conference. I don't know Zep. I think there's a bunch of other memory guys, but like something like this, I think should be normal in the stack. And basically, I think anything stateful should be interesting to VCs. Because it's databases and you know, they, you know how those things make money.

• 16:12Jacob Effron?

  I think on the overhyped side, the only thing I'd add is like, I'm, I'm still surprised how many net new companies there are training models. I thought we were kind of like past that.

• 16:19Speaker 01

  And I would say they died end of last year and now they've resurfaced.

• 16:22Jacob Effron?

  Yeah.

• 16:22Speaker 01

  I mean, that's one of the questions that you had down there of like,

• 16:25Jacob Effron?

  Yeah, sorry.

• 16:25Speaker 01

  Is there an opportunity for net new model players? I would have said no. I don't know what you guys think.

• 16:30Speaker 02

  I, I don't have a reason to say no, but also don't have a reason to say this is what is missing and you should have a new model company do it. But again, I'm an AI engineer.

• 16:39Speaker 01

  All these guys want to pursue AGI. You know, they all want to be like, oh, we'll, we'll like hit, you know, sort of on all the benchmarks. And like, they can't all do it.

• 16:46Jacob Effron?

  Yeah. I mean, look, I don't know if Avilia has the secret, a secret approach up his sleeve of, of something beyond test time compute. But it was funny, I, we had Noam Chizzi on the podcast last week. I was asking him like, you know, is, is there like some sort of other algorithmic breakthrough? What do you make of Avilia? And he's like, look, I think what he implicitly said was test time compute will get us to the point where these models are doing AI engineering for us. And so, you know, at that point, they'll figure out the next algorithmic breakthrough, which I thought was, was pretty interesting.

• 17:11Speaker 03

  I agree with you, Swig, because I think that. We're most interested, at least from our side, in like, you know, foundation models for specific use cases and more specialized use cases. I guess the broader point is if there is something like that, that these companies can latch on to and being there sort of known for being the best at, maybe there's a case for that. Largely, though, I do agree with you that I don't think there should be at this point, more model companies.

• 17:33Jacob Effron?

  I think it's like these unique data sets, right? I mean, obviously robotics has been an area we've been really interested in. It's an entirely different set of data that's required, you know, on top of like a good DLM and then, you know, biology, material science.

• 17:44Speaker 03

  More the specific use cases.

• 17:46Jacob Effron?

  Yeah, but also specific, like, I think a lot of these models are super generalizable, but like, you know, finding opportunities to, you know, where, you know, for a lot of these bio companies, they have wet labs, like they're like running a ton of experiments or, you know, same on the material sciences side. And so I still feel like there's some, some opportunities there, but the core kind of like LLM agent space is tough to compete with the big ones.

• 18:05Speaker 02

  Yeah, agree. Yeah, but they're moving more into product. So I think that's the question is like, if they could do better vertical models, why not do that instead of trying to do deep research? And operator and these different things. I think that's what I'm, in my mind. Sweet agents coming out too. Well, yeah, in my, in my mind, it's like financial pressure, like they need to monetize in a much shorter timeframe because the costs are so high. But maybe it's like, it's not that easy to do.

• 18:32Jacob Effron?

  You think they would be, that it would be a better business model to like do a bunch of vertical?

• 18:35Speaker 02

  It's more like, why wouldn't they? You know, like you make less enemies if you're like a model builder, right? Like, like now with deep research and like search, now perplexities like anatomy and like, you know, Gemini D pre research is like more of an enemy versus if they were doing a finance model. You know, or whatever, like they would just enable so many more companies and they always have, like they had Habia as one of the customer case studies for GPT search. But they're not building a finance based model for them. So is it because it's super hard and somebody should do it? Or is it because the new models are going to be so much better that like the vertical models are useless anyways? This is a bitter lesson. Exactly.

• 19:14Jacob Effron?

  It still seems to be a somewhat outstanding question. I'd say like all the signs of last few years seem to be like a general purpose model is like the way to go. And, you know, you know, like training a hyper specific model in a, in a domain is like, you know, maybe it's cheaper and faster, but it's not going to be like higher quality. But also like, I think it's still, I mean, we're talking to, Noam and Jack Ray from Google last week and they were like, yeah, this is still an outstanding, like we, we check this every time we have a new model, like whether there's, you know, that still seems to be holding. I remember like a few years ago, it felt like all the rage was like, it was like the Bloomberg GPT model came out and everyone was like, oh, you gotta like.

• 19:46Speaker 01

  Yeah, I had the GPT at Bloomberg present on that.

• 19:49Jacob Effron?

  Yeah, that must be a really interesting episode to go back on because I feel like, like very shortly thereafter, the next opening I model came out and just like beat it on all sorts of the.

• 19:56Speaker 01

  Uh, no, it was a talk. We haven't released it yet, but yeah, I mean, it's, uh, basically they concluded that the, the closed models were better. So they just.

• 20:03Jacob Effron?

  Okay, yeah, it's funny, though. Interesting. Yeah, I feel like that's been the, uh.

• 20:06Speaker 01

  Uh, but he's, uh, I, I would be, he's very insistent that the work that they did, the team he assembled, the data that he collected is actually useful for more than just the model. So like basically everything but the model survived.

• 20:17Jacob Effron?

  What are the other things?

• 20:18Speaker 01

  Uh, the data pipeline, the team that they, they assembled for like fine tuning and implementing, uh, whatever models they, they ended up picking. Um, yeah, it seems like they are happy with that and they're running with that. Uh, he runs like 12, 13 teams at Bloomberg, just working Jenny I across the company.

• 20:36Jacob Effron?

  I mean, I guess we've, we've all kind of been alluding to it right now, but I guess because it's a natural transition, you know, the other broad opening I have is just what we're paying most attention to right now. And I think back on this, like, you know, the model companies coming into the product area. I mean, I think that's going to be like, I'm fascinated to see how that plays out over the next year and kind of these like frenemy dynamics and it feels like it's going to first boil up on like cursor anthropic and like the way that plays out over the next six months, I think will be.

• 20:59Speaker 01

  What is cursor anthropic? You mean cursor versus anthropic?

• 21:03Jacob Effron?

  Yeah, I mean, I assume, you know, over time, anthropic wants to get more into the application side of coding. Um, and, you know, I assume over time cursor will want to diversify off of, you know, just using the anthropic model.

• 21:14Speaker 01

  It's interesting that now cursor is now worth like 10 billion, nine, nine, 10 billion.

• 21:18Jacob Effron?

  Yeah.

• 21:18Speaker 01

  Um, and like they've made themselves hard to acquire. Like I would have said, like, you should just, you get yourself to five, six billion and join OpenAI and like all the training data goes to OpenAI and that's how they train their coding model. Uh, now it's, now it's complicated. Now they need to be an independent company.

• 21:33Jacob Effron?

  Uh, increasingly it seems that model companies want to get into the product layer. And so seeing over the next six, 12 months, just having the best model, you know, uh, let you kind of start from a cold start on the product side and get something in market or are the, you know, companies with the best products, even if they eventually have to switch to a somewhat worse, tiny bit worse model, does it not, you know, where do the developers ultimately choose to go? I think that'll be super interesting.

• 21:55Speaker 01

  Yeah.

• 21:55Speaker 02

  But don't you think that Devon is more in trouble than cursor? I feel like Anthropic, if anything, wants to move more towards, I don't think they want to build the IDE. Like, if I think about coding, it's like, kind of like, you know, you look at it like a cube, it's like, the IDE is like one way to get the code and then the agent is like the other side.

• 22:10Speaker 01

  Yeah.

• 22:10Speaker 02

  I feel like Anthropic wants more be on the agent side and then hand you off the cursor when you want to go in depth. Versus like trying to build the Claude IDE, I think that's not.

• 22:21Speaker 01

  I would say the existence of Claude code doesn't show, doesn't support what you say, like, maybe they would, but.

• 22:29Jacob Effron?

  I assume all this, like, I assume both just converge eventually where you want to have, uh, where you'll be able to be both.

• 22:34Speaker 01

  So, so in order to be, uh, so we're, we're talking about coding agents, whether it's sort of, um, what is it, inner loop versus outer loop, right? Like, inner loop is inside cursor, inside your IDE between, uh, inside of a git commit and outer loop is between git commits on, on the cloud. And I think like to be an outer loop coding agent, you have to be more of a, like, we will integrate with your code base, we'll sign your, whatever, you know, security thing that you need to sign. That kind of schlep. Uh, I don't think the model labs want to do that schlep, they just want to provide models. So that, that's, that would be my argument against, like, why cognition should still, uh, have, have, have some moats against Anthropic, just simply because they would. Cognition would do the schlep and the biz dev and the infra that Anthropic doesn't really care about.

• 23:20Jacob Effron?

  I know, the schlep is pretty sticky though, once you do it.

• 23:22Speaker 01

  It's very sticky. Yeah, yeah. Um, I mean, it's, it's, it's interesting, like, I, I think the natural winner of that should be SourceCraft.

• 23:28Jacob Effron?

  But like, there's. Another unprompted, uh.

• 23:33Speaker 01

  I mean, they're, they're big supporters, uh, like, I'm very friendly with both Quinn and Bian, and, uh, they've, uh, they've done a lot of work with Cody. But like, no, not much work on the outer loop stuff yet, but like, any company where like, they have already had, like, we've been around for 10 years. We, we like have all the enterprise contracts, you already trust it to your code base, why would you go trust like Factory or Cognition? As like, you know, two-year-old startups who are like, just came out of MIT, like, I don't know.

• 24:00Jacob Effron?

  I guess switching gears to the, uh, to the application side, um, I'm curious, uh, for both of you, like, how do you kind of characterize what has genuine product market fit in AI today? And, uh, I guess I'll ask you more, and you're sort of on the investing side, like, more interested to invest in that category of the stuff that works today? Or, uh, kind of where the capabilities are going long-term?

• 24:20Speaker 02

  That's hard.

• 24:20Jacob Effron?

  I'm not gonna do my job for it.

• 24:21Speaker 02

  You were like, man, that's a easy, that's a layup.

• 24:24Speaker 01

  Tell us all your investment pieces.

• 24:25Speaker 02

  Yeah. I would say, well, we only really do mostly seed investing. So it's hard to invest in things that already work. Because it means they're already late. So we try to, but we try to be at the cost of like, you know, usually the investments we like to make, there's like really not that much market risk. It's like, if this works, obviously people are gonna use it. But like, it's unclear whether or not it's gonna work. So that's kind of more what we see towards. We try not to chase as many trends. And I don't know, you know, I was a founder myself. And sometimes I feel like it's easy to just jump in and do the thing that is hot. But like, becoming a founder to do something that's like underappreciated, or like doesn't yet work, shows some level of like grit and self, like you actually really believe in the thing.

• 25:10Speaker 02

  So that alone for me is like, kind of makes me feel more towards that. And you do a lot of angel investing too, so I'm curious how.

• 25:16Speaker 01

  Yeah, but I don't regard, I don't have, I don't use, put that in my mental framework of things. Like, I come at this much more as a content creator or market analyst of like, yeah, it really does matter to me what has product market fit. Because people, I have to answer the question of what is working now when people ask me.

• 25:35Jacob Effron?

  Do you feel like relative to the, obviously the hype and discourse out there, like, you know, do you feel like there's a lot of things that have product market fit? Or like a few things, like where?

• 25:43Speaker 01

  A few things.

• 25:44Jacob Effron?

  Yeah.

• 25:44Speaker 01

  So I have a list of like, two years ago, I wrote the Anatomy of Autonomy post, where it was like the first like, what's going on in agents? And, and, and what is actually making money? Because I think there's a lot of Gen I skeptics out there. They're all like, these, these things are toys, they're unreliable. And, you know, why, why are you dedicating your life to these things? And I think for me, the product market fit bar at the time was a hundred million dollars. Right? Like what use cases can reasonably fit a hundred million dollars? And at the time, it was like Copilot, it was Jasper, no longer. But, you know, in that category of like, I'll help you write.

• 26:24Jacob Effron?

  Yeah.

• 26:24Speaker 01

  Which I think was helpful. And then, and Cursor, I think was on there as, as, as, as like a coding agent, uh, plus plus. I think that list will just grow over time of like the form factors that we know to work. And then we can just adapt the form factors to a bunch of other things. So like the, the one that's the most recently added to this is deep research.

• 26:44Jacob Effron?

  Yeah.

• 26:44Speaker 01

  Right, where anything that looks like a deep research, whether it's a Grok version, Gemini version, perplexity version, whatever, he has an investment that he likes called Brightwave that is basically deep research for finance.

• 26:56Jacob Effron?

  Yeah.

• 26:56Speaker 01

  And anything where like, all right, it's like long-term agentic reporting is starting to take more and more of the job away from you and just give you a much more reason to report, I think it's going to work and that has some PMF, I think, obviously has PMF. Like, I would say it's, I went to this exercise of trying to handicap how much money OpenAI made from launching OpenAI Deep Research, I think it's billions. Like, the share upgrade from like $20 to $200, it has to be billions in the era. Maybe not all of them will stick around, but like, that is some amount of PMF that is...

• 27:34Jacob Effron?

  Didn't they have to immediately drop it down to the $20 tier?

• 27:36Speaker 01

  They expanded access. I don't, I wouldn't say...

• 27:38Jacob Effron?

  Which I thought was really telling of the market, right? It's like where you have a, you know, I think it's gonna be so interesting to see what they're actually able to get in that $200 or $2,000 tier, which we all think is, you know, has a ton of potential, but I thought it was fascinating. I don't know whether it was just to get more people exposure to it or the fact that like, Google had a similar product, obviously, and other folks did too. But it was really interesting how quickly they dropped it down.

• 27:58Speaker 01

  I don't, I think that's just a more general policy of no matter what they have at the top tier, they always want to have smaller versions of that in the, in the lower tiers.

• 28:07Jacob Effron?

  Yeah, just get people exposure to it.

• 28:08Speaker 01

  Yeah, just get exposure. The brand of being first to market and, and like the default choice is paramount to OpenAI.

• 28:16Jacob Effron?

  Though I thought that whole thing was fascinating because Google had the first product, right?

• 28:18Speaker 01

  Yeah.

• 28:19Jacob Effron?

  And no, like, you know.

• 28:21Speaker 01

  We interviewed them. I, I straight up to their faces, I was like, "OpenAI mocked you." And they were like, "Yeah."

• 28:25Jacob Effron?

  Well, I actually, this is totally off topic, but whatever. Like, what is it going to take for Google just release some great models like a few weeks ago? I feel like...

• 28:34Speaker 01

  It's happening.

• 28:34Jacob Effron?

  The stuff they're shipping is really cool.

• 28:36Speaker 01

  It's happening.

• 28:37Jacob Effron?

  Um, yeah. But I also, I feel like at least in the, you know, broader discourse, it's still like a drop in the bucket relative to...

• 28:43Speaker 01

  Yeah, I mean, I can riff on this, but I think it's happening. I think it takes some time. But I am like my Gemini usage is up. Like I use I use it a lot more for anything from like summarizing YouTube videos to the native image generation that they just launched to like flash thinking.

• 29:02Jacob Effron?

  Multi-level stuff is great.

• 29:03Speaker 01

  Yeah, I run, you know, and I run like a daily sort of news recap called the I news that is 99% generated by models. And I do a bake off between all the frontier models every day. And it's

• 29:14Jacob Effron?

  Every day, it's like, does it switch?

• 29:16Speaker 01

  I manually, yes, it does switch and I manually do it. And flash is flash wins most days.

• 29:24Speaker 01

  So so like, I think it's happening. I think I was thinking I was thinking about tracking myself. Like number of opens of chat GPT versus Gemini. And at some point it will cross. I think that Gemini will be my main. And I like that will slowly happen for a bunch of people and then that'll shift. I think that's that's a really interesting for developers. This is a different question. It's Google getting over itself of having Google Cloud versus Vertex versus AI Studio. They're all these like five different brands slowly consolidating it. It'll happen just slowly, I guess.

• 29:58Speaker 02

  Yeah, I mean, another good example is like you cannot use the thinking models in Cursor. And I know Logan Kielpatrick said they're working on it, but I think there's all these small things where like, if I cannot easily use it, I'm really not going to go out of my way to do it. But I do agree that when you do use them, their models are are great. So yeah, they just need better better bridges.

• 30:20Speaker 01

  You had one of the questions in the prep, what public company are you long and short? And mine is Google versus versus Apple.

• 30:26Jacob Effron?

  That was also my combo. I feel like, yeah, I mean, it does feel like Google is really cooking right now.

• 30:31Speaker 01

  Yeah. So, okay, coming back to what is product market fit?

• 30:36Jacob Effron?

  Now that we come back to my complete total sidetrack.

• 30:39Speaker 01

  There's also customer support. We were talking on the car about Decagon and Sierra. Obviously Brett Taylor is founder of Sierra. And yeah, it seems like there's just these layers of agents that'll like, I think you just look at like the income statement or like the org chart of any large scaled company and you start picking them off one by one. What like is interesting knowledge work? And they would just kind of eat things slowly from the outside in. Does that make sense?

• 31:05Speaker 02

  I mean, the episode with Brett, he's so passionate about developer tools. And yeah, he did not do a developer tools conference. We spent like two hours talking about developer tools and like all of that stuff. And he's like, I did a customer support company. And I'm like, man, that says something. You know what I mean? It's like when you have somebody like him who can like raise any amount of money from anybody to do anything.

• 31:26Speaker 01

  Yeah.

• 31:27Speaker 02

  To pick customer support as the markets to go after while also being the chairman of OpenAI. Like, that shows you that like, these things have modes and have longstanding, like they're gonna stick around, you know, otherwise he's smarter than that. So yeah, that's a, that's a space where maybe initially, you know, I would have said, I don't know, it's like the most exciting thing to to jump into. But then if you really look at the shape of like, how the workforce are structured and like, how the cost centers of like the business really end up, especially for more consumer facing businesses, like a lot of it goes into customer support. All the AI story of the last two years has been cost cutting. I think now we're gonna switch more towards growth.

• 32:07Speaker 01

  Revenue.

• 32:08Speaker 02

  You know, like you've seen Jensen, like last year at GTC was saying, the more you buy, the more you save. This year he said, the more you buy, the more you make. So we're.

• 32:17Jacob Effron?

  Hot off the press.

• 32:18Speaker 02

  We were there, we were there.

• 32:19Jacob Effron?

  I do think that's one of the most interesting things about this first wave of apps where it's like almost the easiest thing that you could, you could get real traction with was stuff that, you know, for lack of a better way to frame it, like some of the people had already been comfortable outsourcing to BPOs or something and kind of implicitly said like, hey, this is a cost center. Like, we are willing to take some performance cut for cost in the past. You know, the irony of that, or what I'm really curious to see how it plays out is, you know, you could imagine that is the area where price competition is going to be most fierce, because it's already stuff that, you know, that people have said, hey, we don't need the like 100% best version of that. And I wonder, you know, this next wave of apps may prove actually even more defensible as you get these capabilities that actually are, you know, increase top line or whatnot, where you're like, you take AI go to market, for example, like you're, you pay like twice

• 33:04Jacob Effron?

  as much for something that brought like, because there's just a kind of very clean ROI story to it. And so I wonder ultimately whether the like this next set of apps actually ends up being more interesting than the than the first wave.

• 33:16Speaker 02

  Yeah.

• 33:16Speaker 03

  I think a lot of the voice AI ones are interesting too, because you don't need 100% precision recall to actually, you know, have a great product. And so, for example, we looked into a bunch of, you know, scheduling intake companies, for example, like home services, right, for electricians and stuff like that. Today, they miss 50% of their calls. So even if the AI is only effective, say, 75% of the time, yeah, it's crazy, right? So if it's effective 75% of the time, that's totally fine, because that's still a ton of increased revenue for the customer, right? And so you don't need that 100% accuracy. And so as the models and the reliability of these agents are getting better, it's totally fine, because you're still getting a ton of value in the meantime.

• 33:53Speaker 01

  Yeah, what this is, I don't know how related this is, but I'm one of my favorite meetings at, it is related.

• 34:01Speaker 01

  One of my favorite meetings at AI engineer summit, because like I do these, this was our first one in New York, and I just like met the different crew than than you meet here. Like, everyone here is loves developer tools, loves infra. Over there, they're actually more interested in applications, it's kind of cool. I met this like bootstrap team that like they're only doing appointment scheduling for vets. And like, they're like, this is a, this is an anomaly, we don't usually come to engineering summits, because we usually go to vet summits. And like, talk to the, they're like, you know, they're, they're literally the.

• 34:29Speaker 03

  I'm sure it's a massive pain point, they're willing to pay a lot of money.

• 34:33Speaker 02

  But but this is like my point about saving versus making more. It's like if an electrician takes two X more calls, do they have the bandwidth to actually do two X more in-house? Well, yeah, exactly. That's the thing. It's like, I don't think today, most businesses are like structured to just like overnight two, three X the band, you know? I think that's like a startup thing, like mo- most businesses.

• 34:54Speaker 01

  Then you make an electrician agent.

• 34:56Speaker 02

  Well, not totally. So how do you do? How do you do a recruiting agent for an electrician? But like, electrician training, how do you do Lambda School for electrician? I don't know. It's like.

• 35:07Jacob Effron?

  Whack a mole for the bottlenecks in these businesses, like as.

• 35:09Speaker 02

  Yeah, exactly.

• 35:10Jacob Effron?

  Now we have a ton of demand, like, cool, where do we go?

• 35:12Speaker 01

  Yeah. So just to round out the the PMF thing, I think this is relevant in a sense of like, it's pretty obvious that the killer agents are coding agents, support agents, deep research, right? Roughly, right? We've covered all those three already. Then then you have to sort of be, go turn to offense and go like, okay, what's next?

• 35:30Jacob Effron?

  And like, what about, I mean, I also just like summarization of of voice and conversation.

• 35:33Speaker 01

  Yep. Well, we actually had that on there. It's I just, I didn't put it as agent, because it seems less agentic, you know? But yes.

• 35:40Jacob Effron?

  Still a good AI use case.

• 35:41Speaker 01

  Uh, that one I I've seen, I would mention granola and what's the other one? Uh, Monterey?

• 35:47Jacob Effron?

  I think a bridge was.

• 35:48Speaker 02

  I was gonna say a bridge, yeah.

• 35:49Speaker 01

  Bridge, okay. So I'll just, I would call out what I had on my slides for for the agent engineering thing. Um, so it was screen sharing, which I think is actually a kind of underrated, like people like, yeah, watching you as you do your work can just think offering assistance, outbound sales. So instead of support, just being more outbound, hiring.

• 36:08Jacob Effron?

  You say outbound sales has product market fit?

• 36:10Speaker 01

  No, it it will, it's coming.

• 36:11Jacob Effron?

  Oh, on the come, yeah, I was totally agree with that.

• 36:13Speaker 01

  Yeah, um, hiring, um, like the recruiting side, uh, education, like the the sort of like personalized teaching, I think.

• 36:20Jacob Effron?

  Um, I'm kind of shocked we haven't seen more there.

• 36:23Speaker 01

  Yeah.

• 36:23Jacob Effron?

  I don't know if that's like.

• 36:24Speaker 01

  It's like Duolingo is the thing, Camigo.

• 36:27Jacob Effron?

  Yeah, I mean, speak in some of these like, you know, practice.

• 36:29Speaker 01

  Yeah, yeah, interesting. And then finance, there's a, there's a ton of finance cases that we can talk about that. And then personal AI, which we also had a little bit of that. But I think personal AI is harder to monetize. But I think those would be like, what I would say is up and coming in terms of like, that's what I'm currently focusing on.

• 36:47Jacob Effron?

  I thought this question has been asked a few different ways. But I'm curious what you guys think it's like. Is it like, if we just froze model capabilities today, like, is there, you know, trillions of dollars of application value to be unlocked? Like, like AI education, like if we just stopped today, all model development, like, with this current generation of models, we could probably build some pretty amazing education apps. Or like, how much of this, how much of all this is like contingent upon just like, okay, people have had two years with GPT-4 and like, you know, I don't know, six months with the reasoning models? Like, how much contingent upon it just being more time with these things versus like, the models actually have to get better? I don't know, it's a hard question, so I'm gonna just throw it to you.

• 37:21Speaker 02

  Yeah. Well, I think the societal thing is maybe harder, especially in education. You know, like, can you basically like Doge the education system? Probably you should, but like, can you? I, I think it's more of a human.

• 37:36Jacob Effron?

  But people pay for all sorts of like, get ahead things outside of class. And, you know, certainly in other countries, there's a ton of consumer spend in education. It feels like the market opportunity is there.

• 37:44Speaker 01

  Yeah, and in private education, I think. Yeah, public, public is very different. One of my most interesting quests from last year was kind of reforming Singapore's education system to be more sort of AI native.

• 37:56Jacob Effron?

  Just what you were doing on the side while you're.

• 37:58Speaker 01

  Yes.

• 37:58Jacob Effron?

  That's a great, that's a great side quest.

• 38:01Speaker 01

  My stated goal is for Singapore to be the first country that has Python as the first language, as a, as a national language. Anyway, so, but the, the, the defense, the pushback I got from Ministry of Education was that the teachers would be unprepared to do it. So it was like, it was like the, like the, it was really interesting, like immediate pushback was the de facto teachers' union being like resistant to change. And I'm like, okay, that, that's par for the course. Anyway, so not, not to, not to dwell too much on that, but like, yeah, I mean, like, I, I think like, education is one of these things that everyone like has strong opinions on, because they all have kids, all enter the education system. But like, I think it's gonna be like the, the domain specific, like, like speak, like,

• 38:43Speaker 01

  such an amazing example of like, top down, like, we will go through the IDMA's and we'll go to Korea and teach them English. Like, it's like, what? And I would love to see more examples of that, like just like really focus, like no one try to solve everything. Just just do your thing really, really well.

• 39:00Jacob Effron?

  On this trend of difficult questions that come up, I'm gonna just ask you the one that my partners like to ask me every single Monday, which is, how do you think about defensibility at the at the app layer?

• 39:08Speaker 01

  Oh, yeah, that's great.

• 39:10Jacob Effron?

  Just give me an answer, I can copy paste and just like, you know, auto response.

• 39:14Speaker 01

  Honestly, like network effects, I think people don't prioritize those enough because they're trying to make the single player experience good, but then then they neglect the multiplayer experience. I think one of the, I always think about like load bearing episodes, like, you know, as as parts that you do one a week and like, you know, some of those you don't really talk about ever again. And others you keep mentioning every single podcast and.

• 39:38Jacob Effron?

  This is obviously gonna be the last.

• 39:41Speaker 01

  I think the recap episodes for us are pretty load bearing, like we we refer to them every three months or so. And like, one of them, I think for for us is Chai, for me is Chai Research, even though that wasn't like a super popular one among the broader community outside of Chai, the Chai community. For those who don't know, Chai Research is basically a character AI competitor, right? They were bootstrapped, they were founded at the same time, and they have outlasted character de facto, right? It's funny, like I would love to ask no one, she's there a bit more about like the whole character thing, but.

• 40:11Jacob Effron?

  Good luck getting past the Google.

• 40:15Speaker 01

  But like, so he like, he like, he doesn't have his own models, basically. He has his own network of people submitting models to be run. And I think like, that is like short term going to be hurting him because he doesn't have proprietary IP. But long term, he has the network network effect to make him robust to any changes in the future. And I think like, I want to see more of that, where like he's basically looking at himself as kind of a marketplace and he's identified the choke point, which is both the app or the the sort of protocol layer that interfaces between the users and the model providers, and then make sure that the money kind of flows through. And that works. I wish that more AI builders or AI founders

• 40:57Speaker 01

  emphasize network effects, because that that's the only thing that you're gonna have at the end of the day. And like brand deeds into network effects.

• 41:04Jacob Effron?

  Yeah, I guess harder in the enterprise context, right? But I mean, I feel it's funny, we do this exercise and I feel like we talk a lot about like, you know, obviously there's, you know, kind of the velocity and the breadth you're able to kind of build of product surface area. There's just like the ability to become a brand in a space. Like I'm shocked that even in like six, nine months, how an individual company can become synonymous with like an entire category and like then they're in every room for customers and like all the other startups are like clawing their way to try and get in like one, you know, twentieth of those rooms.

• 41:31Speaker 03

  There's a bunch of categories where we talk about in IC and it's like, oh, pricing compression is going to happen, not as defensible, and so ACVs are going to go down over time. In actuality, some of these, the ACVs have doubled, we've seen. And the reason for that is just, you know, people go to them and pay for that premium of being that brand.

• 41:48Jacob Effron?

  Yeah, I mean, what I'm struck by is there's been, there was such a head fake in the early days of AI apps where people were like, we want this amazing defensibility story. And then what's the easiest defensibility story? It's like, oh, like totally unique data set or like train your own model or something. And I feel like that was just like a total head fake where I don't think that's actually useful at all. It's the much less, you sound much less articulate when you're like, well, the defensibility here is like the thousand small things that this company does to make like the user experience, design, everything just like delightful and just like the speed at which they move to kind of both create a really broad product, but then also every three, six months when a new model comes out is kind of an existential event for like any company. Cause if you're not the first to like figure out how to use it, someone else will. And so velocity really matters there. And it's funny, in kind of our internal discussions, we've been like, man, that sounds pretty similar to like how we thought about like application SaaS companies.

• 42:32Jacob Effron?

  But there isn't some like revolutionary reason, you don't sound like a genius when you're like, here's applications, why application SaaS company A is so much better than B. But it's like a lot of little things that compound over time. What about the infrastructure space, guys? Like, I'm curious, you know, how do you guys think about where the interesting categories are here today? And, you know, like where, where, where do you want to see more startups or where do you think there are too many?

• 42:54Speaker 02

  Yeah, we call it kind of the L L M O S. But I would say.

• 42:57Speaker 01

  No, we, I mean, Andre, Andre calls it L M O S.

• 43:00Speaker 02

  Well, but yeah, well.

• 43:03Jacob Effron?

  You, you and Andre, the three of you call it the L M O S.

• 43:05Speaker 02

  Well, we have this like four words of AI framework that we use and L M O S is one of them. But yeah, I mean, code execution is one, we've been banging the drum. Everybody now knows where investors in E to P. Memory, you know, is one that we kind of touched on before. Super interesting. Search, we talked about, I think those are more. Not traditional infra, not like the bare metal infra. It's more like the infra around the model, you know? Uh, which I think is where a lot of the value is gonna be. Um, the Apple. Security ones? Yeah, yeah, and cyber security, I mean, um, there's so much uh-huh to be done there. And it's more like basically any area where AI is being used by the offense.

• 43:47Speaker 02

  AI needs to be applied on the defense side, like email security, you know, identity, like all these different things. Um, so we've been doing a lot there a- as well as, you know? How do you rethink things that used to be costly like red teaming and maybe used to be a checkbox in the past? Today, they can be actually helpful uh to make you secure your app. And there's the sole idea of like semantics, right? That not the models can be good at. Um, you know, in the past, everything is about syntax. It's kind of like very basic, you know, constraint rules. I think now you can start to infer semantics from things that are beyond just like simple recognition to like understanding why certain things are happening a certain way. So in the security space, we're seeing that with um binary inspection, for example, like

• 44:32Speaker 02

  there's kind of like the syntax, but then there are like semantics of like understanding what is the scope overall really trying to do, even though this individual syntax is like seeing something specific. Not to get too technical, but um, yeah, I I think infra overall is like a super interesting place if you're making use of the model. If you're just, I I'm less bullish, not not that it's not a great business, but I think it's a very capital intensive business, which is like serving the models. I think that impress like great people will make money, but um, yeah, I I I don't think there's as much of a um interest from from us.

• 45:08Speaker 03

  Yeah. How how do you guys think about what OpenAI and the big research labs will encompass as part of the developer and infra category?

• 45:15Speaker 02

  Yeah, that that's why I I would say. Um. Search is the first example of one of the things we used to mention on, you know, we had X on the podcast and perplexity, obviously, as an API.

• 45:27Speaker 01

  The basic idea is if you go into like the chat GPT custom GPT builder, like what are the checkboxes? Each of them is a startup.

• 45:33Speaker 02

  Yeah. And and now they're also APIs. So now search is also an API.

• 45:39Speaker 02

  We'll see what the adoption is. There's the, you know, in traditional infra, it's like everybody wants to be multi-cloud. So maybe we'll see the same where um chat GPT search or OpenAI search API is like great with the OpenAI models because you get it all bundled in. But their price is very high. If you compare it to like, you know, X, I think it's like five times the the price for the same amount of research, which makes sense if you have a big OpenAI contract, but maybe if you're just like pick and best and breed, you want to compare different ones.

• 46:08Speaker 01

  Yeah.

• 46:09Speaker 02

  Um. Yeah, they don't have a code execution one. I'm sure they'll release one soon. Um. So they want to own that too, but um. Yeah. The same question we were talking about before, right? Did they want to be an API company or a product company? Do you make more money building chat GPT search or selling search API?

• 46:24Speaker 01

  Yeah, the the broader lesson, instead of like going, we did applications just now, and then what do you think is interesting infrastructure? Like, it's not fifty fifty. It's not like equal weighted. Like, it's just very clearly the application layer has like been way more interesting.

• 46:38Speaker 01

  Like, yes, there is interesting infrastructure plays, and I even want to like push back on like the the the whole GPU serving thing. Because like together, AI is doing well, fireworks.

• 46:47Jacob Effron?

  I mean, that's that's what it's like. It's like data centers and inference providers.

• 46:49Speaker 02

  Yeah, yeah.

• 46:50Jacob Effron?

  You know.

• 46:51Speaker 02

  I think it's all like the capital.

• 46:52Speaker 01

  Ah, yes, yes.

• 46:53Speaker 02

  For for again.

• 46:55Speaker 01

  Capital efficiency.

• 46:56Speaker 02

  Much larger funds, so you know, I'm sure you have GPU clouds.

• 46:59Speaker 01

  Yeah. Yeah. So um, that's that. That is one thing I have been learning in in that, um, you know, I think I have historically had DevTools and Infra Bias. And so is he. And we've had to learn that applications actually are very interesting and also maybe kind of the killer application of models in the sense that you can charge for utility and not for cost, right? Which where like most infrastructure reduces to cost plus.

• 47:24Speaker 02

  Yeah.

• 47:25Speaker 01

  Right? So and like, that's not where you want to be for AI.

• 47:29Speaker 01

  So that's that's interesting for for me. I thought it would be interesting for me to be the only non VC in the room to be saying what is not investable. Because like then I then, you know, you can, I won't be canceled for saying like your your whole category.

• 47:43Jacob Effron?

  This thing is not investable and then like three months later, we're desperately chasing.

• 47:45Speaker 01

  Exactly. So you don't want to be on the record.

• 47:48Jacob Effron?

  It changes so fast. Yeah, every opinion you hold, you have to like hold it quite loosely.

• 47:51Speaker 01

  I'm happy to be wrong in public, you know, I think that's how you learn the most, right? So like fine tuning companies is something I struggled with and still like I don't see how this becomes a big thing. Like you kind of have to wrap it up in a broader broader enterprise AI company, like services company, like a writer AI. Where like they will find you and it's part of the overall offering, but like that's not where you spike. Yeah, it's kind of interesting. And then I'll just kind of AI DevOps and like there's a lot of AI SRE out there. Seems like there's a lot of data out there that that should be able to be plugged into your code base or or your app to it. It's sort of self heal or whatever. It's just, I don't know if that's like been a thing yet and you guys can correct me if you're if I'm wrong.

• 48:35Speaker 01

  And then the last thing I'll mention is voice, real time infra. Again, like very interesting, very, very hot, but again, how big is it? Those are the main three that I'm thinking about for things I'm struggling with.

• 48:47Speaker 03

  Yeah, I guess a couple of comments on the AI SRE side, I actually disagree with that one. I think that the reason they haven't sort of taken off yet is because the tech is just not there quite yet. And so it goes back to the earlier question, do we think about investing towards where the companies will be when the models improve versus now? I think that's going to be in short term, we'll get there, but it's just not there just yet. But I think it's an interesting opportunity overall.

• 49:11Speaker 01

  Yeah, it's my pushback to you is well, it's monitoring a lot of logs, right? And it's basically anomaly detection rather than like there's there's a whole bunch of like stuff that can happen after you detect the anomaly. But it's really just not anomaly detection and we've always had that. You know, like it's this is like not a transformers L M use case. This is just regular anomaly detection.

• 49:32Speaker 03

  It's more in terms of like, it's not going to be an autonomous S R E for a while. And so the question is, how how much can the latest sort of A I advancements increase the efficacy of going? Bring your M T T R down.

• 49:44Speaker 01

  Yeah.

• 49:44Speaker 03

  And even if it's 10% improvement on beforehand, that's still potentially a lot of revenue.

• 49:50Speaker 01

  Okay.

• 49:51Speaker 03

  That's the way at least I think I would think about it now. And then, you know, a few years from now, if it's actually an autonomous S R E just replacing altogether, then that's a totally different thing.

• 50:01Speaker 01

  Cool, I'll look out for it.

• 50:03Jacob Effron?

  Yeah. You know, I guess switching back to overly broad questions like, what do you feel like is the biggest unanswered question in A I today? You know, that has and you know, large implications for the ecosystem.

• 50:15Speaker 02

  Yeah, I I've been banging the drum on R L and I think it's clear that you can do R L successfully on verifiable domains. I would say whether or not we can figure out how to do that in non verifiable one. So law is a great example of like, can you do R L on contracts and documents? Marketing, sales, going back to outbound sales, like, can you do R L to like simulate what an outbound? And and kind of like the conversation leads to. Yeah, it's unclear. If not, then I think we'll be stuck with like, you're gonna have agents in the more verifiable domains. And then you'll just kind of have co-pilots in the non verifiable ones because you'll still need a person to be the tastemaker.

• 50:55Jacob Effron?

  I think exactly same thing. And I thought it's like the I just I'm trying to think of the implications where if it doesn't work, like the world could be weird where like you have like fully autonomous A I coders and like, you know, no one does any software or math or even like, you know, some areas of science. But then like to write the most basic sales email is still like, like it's always so hard to predict out the world. Like that is such a weird of all the sci fi that was written, you know, 50 years ago. I don't think anybody foresaw that future. That is a really weird future.

• 51:21Speaker 01

  Did either of you have a different one for that?

• 51:23Speaker 03

  Well, yeah, we'll go back and forth.

• 51:27Jacob Effron?

  Biggest unanswered question, I guess.

• 51:31Speaker 03

  I don't know if this is a good answer, but you know, Bob McGrew we had on the podcast, and he was talking about like the rule of nines they have at OpenAI where to go from 90% reliability to 99, it's an order of magnitude increase in compute, and then 99 to 99.9, order of magnitude increase, and that happens every two to three years. And so I think, how are we going to scale sort of accordingly, this sort of next part? I think there's a lot of unanswered questions, just like from a hardware perspective. And then I think as part of that, from the availability perspective, like is Nvidia just going to continue to be dominant? Like, obviously AWS is going hard into, what's their, Tranium chips? I'm blank on that, thank you. And so, I think like,

• 52:14Speaker 03

  there's a big ecosystem around CUDA that's obviously allowed Nvidia to remain dominant, but just, what's going to happen? And is there anyone that's going to come sort of combat that to increase the availability GPUs? Or are we just going to be constrained going forward, when we actually need way more compute going forward?

• 52:31Speaker 01

  Yeah. My quick thoughts, I've been, I'm the only individual named as an investor in Maddox, which is kind of like really funny, because everyone else has funds and none of us has me.

• 52:45Speaker 01

  And it's, there's an interest, like all these like Nvidia startups, sorry, dedicated silicon startups that are coming up and trying to challenge that. And the simple answer is, like, these GPUs are the most general things possible by design. That's why they do gaming and crypto and AI. And I think as long as the architecture seems stable, it seems like there's a case to be made for that. The only question is, who will win that? And obviously, there's a whole bunch of competitors, including, I think AMD is trying to, you know, to make a play for it. But so will AWS, and so will, you know, every other, like Microsoft has a chip, Facebook has a chip. So who knows who will win that? It's just, it's very interesting that,

• 53:28Speaker 01

  This seems to be such a valuable prize, like it's freaking Nvidia that you're competing with.

• 53:32Jacob Effron?

  Yeah.

• 53:33Speaker 01

  And no one has really made a real dent there yet. But so I kind of, I kind of agree with you, but like, I think that basically it's all about stability of workload. And as long as, it's a bet on like the death of transformers, basically. And if you're fine with that, like even, and I think like the, even a state space model, people would agree that like it wouldn't really change that much. And probably, I think the, the overall consensus is that you don't even use state space models individually. Like you would use them in a mixture with transformers anyway. So then like, yeah, just go bet on transformers, bake it into the chip. And you will have much, way more, you know, basically ASICs like for transformers and that's fine.

• 54:15Speaker 01

  And, and so like, prima facie, there should be a company that wins that. I don't know who will win.

• 54:21Jacob Effron?

  I wish we knew.

• 54:23Speaker 01

  I, I think that, I think anyone, you have to start basically after 2019 to 2020. Because anyone started before that will still be too general. Because you, transformers hadn't won yet, one at the time. I have one more. I think that the most emergent one that came out of the New York conference that I did was agent authentication. I think literally the, like the information just published, like this is something that they're worried about, which is when operator or whoever accesses your website on behalf of you, how does it indicate that it's not you, but it's, it's an agent of you? And I think like, my general philosophy on agent experience or any of the sort of like reinvention of every part of the stack for agents is they're all not, not necessary, except for this agent auth thing.

• 55:07Speaker 01

  Like we, we really need to, to be able to like new, new SSO effectively for agents.

• 55:13Jacob Effron?

  Is it gonna be crypto? Both crypto people are really amped about the.

• 55:15Speaker 01

  You know, like, it really, it's really frustrating when Sam Altman is right. But like, maybe you have to scan your eyeballs. Like, maybe you just have to. You just, like, maybe he saw this like five years ago and he was like, you gotta scan your eyeballs. And like, the rest of us are just behind him as usual.

• 55:33Jacob Effron?

  Well, okay, now I'll move to the quickfire round where we'll go around the horn and get quick takes on things. So the first is going to be a dream podcast guest.

• 55:42Speaker 02

  John Carmack.

• 55:44Speaker 01

  Yeah, he's six, John is like six steps away from solving AGI apparently. So we just ask him how long, how far along he is. For us is Andre, for me is Andre. He had, he's a listener and supporter of the pod. And like, basically, when I launched the whole AI engineer push that we that we have, he was the basically the first one to legitimize it. He was like, you know, there will be more AI engineers than ML engineers. And I think that made everyone else pay attention. So like, Lanespace only exists because, you know, he helps, he and other people help to promote it.

• 56:15Speaker 03

  Yeah, I also had Andrea, so I guess we're thinking the same thing there.

• 56:19Jacob Effron?

  I basically, mine's a little bit cheating, but I think at some point there will, like, they're writing a book about OpenAI now. And at some point, like, somebody probably acquired, we'll get to do the acquired OpenAI episode. But if unsupervised learning could like, there's clearly just like so many amazing stories of like, the last five, six years.

• 56:35Speaker 01

  So do you know about Doomers? It's a play. I'm actually going to it this Saturday. Yeah, they, someone made a play about the board drama of last year.

• 56:44Jacob Effron?

  Really?

• 56:44Speaker 01

  Of like, two years ago. Yeah.

• 56:46Jacob Effron?

  Wow. That's cool. I'll tell you how it is.

• 56:48Speaker 02

  Yeah, let us know.

• 56:48Speaker 01

  There will be, there will be lots.

• 56:50Jacob Effron?

  Maybe we should have the director of that on the pod.

• 56:52Speaker 01

  No, I think it's a lot of fanfiction, basically. But like, someone will write the accounts and it will be interesting and fascinating. And a lot of, a lot of it will be fake, because it's a complex beast, right? You're just getting an oral history of what happened.

• 57:05Jacob Effron?

  Yeah.

• 57:05Speaker 01

  Yeah.

• 57:07Jacob Effron?

  All right. For the next one, I figured you could shout out either like, a news source you use to stay up to date, or a startup that's that you're not invested in, that you're excited about.

• 57:14Speaker 01

  Oh.

• 57:15Speaker 02

  Or you can do it. My news source is Sean.

• 57:18Speaker 03

  That's what I was gonna say. I literally wrote Swyx's Twitter.

• 57:22Speaker 02

  In our, in our, we have a Lanespace, we have a Lanespace Discord. Any link that ever matters on the internet, Sean is gonna post it in the Discord. So all I do, I open Discord and we have like, you know, forty, fifty different channels by topic. I open Discord and I'm like, okay, AI, then I go developer tools, then I go creator economy, then I go stock and macro, then I go. And they're all there, so thank you.

• 57:48Speaker 01

  We actually met because of the Discord. It was like a COVID thing because everyone's at home and just started a Discord. And yeah, that was the origin of LeanSpace, just chatting on the Discord.

• 57:57Speaker 02

  It used to be called Dev slash Invest.

• 57:59Speaker 01

  Yeah.

• 57:59Speaker 02

  So it was all about developer tools investing. And then we were OpenAI in October of twenty twenty two. We're like, maybe we should do a podcast. And then OpenAI was the first guest.

• 58:10Speaker 01

  Yeah, I was not prepared about the news sources thing. I think maybe it's hard, it's really shitty to say, but like just in-person conversations.

• 58:23Jacob Effron?

  Yeah.

• 58:24Speaker 01

  And I think the reason I have to be here in SF is because I make friends with people who know things and are smarter than me. And we go for chats and they're nice enough to share some stuff. And so sometimes I wish, I worry that I am being used in order to put things out there that are maybe not true. But, you know, so I have to exercise my own judgment as to what.

• 58:48Jacob Effron?

  I think one of the cool things about the podcast in general is just like the opportunity to take these conversations that happen in like closed rooms and try and bring them on to the airwaves. I'm curious, like, how much of what you, how much do you feel like the private discourse is similar to the, to the public discourse?

• 59:03Speaker 01

  In many ways, it is surprisingly similar.

• 59:07Jacob Effron?

  Yeah.

• 59:08Speaker 01

  As in, people at OpenAI learn about things about OpenAI from us, which is interesting. And then there are some ways in which it is drastically not, drastically dissimilar. And those are the things I just cannot repeat until it's public.

• 59:22Jacob Effron?

  This has been super fun. I feel I've lived up to it. We were looking forward to this for a while. We want to make sure everyone around the horn gets an opportunity to plug whatever they want to plug. So we'll leave the last word to all of us, I guess. Where can folks go to learn more about LeanSpace and all the exciting things you do? Want to make sure our listeners have a good sense of everything.

• 59:40Speaker 02

  Yes, so we have a sub stack, Late in the Space is the website. And then please subscribe on YouTube. Uh, we're doing a lot on YouTube. We're trying to do better video and all that.

• 59:50Speaker 01

  He said our OKRs and it's basically all YouTube.

• 59:53Speaker 02

  Come, come watch us on YouTube. It's very important for me personally. Even if you don't care, just open it.

• 59:59Jacob Effron?

  Gotta hit those OKRs.

• 1:00:00Speaker 01

  Well, we have to increase our production value. Look at this.

• 1:00:03Speaker 02

  I know. We only have three cameras.

• 1:00:07Speaker 02

  Yeah, and then Sean does a lot of the writing outside of the podcast on the newsletter. So.

• 1:00:13Speaker 01

  Yeah, so it's like trying to be a newsletter and community and podcasts and whatever else that we do. Yeah, so I guess for for me, I guess there's Late in Space, but then there's also the other big piece is just the the conference that I run. Um, and the idea is that I think sometimes you just get the the good stuff from people if you just put them in front of a lot of people. Uh, and that's really like I'm mining people for content and sometimes you put a mic in front of them and they yap for an hour. Other other times you have to put them in front of like a prestigious conference and then they drop some alpha. And so the next one for us is going to be June. It's the AI Gartner World's Fair. Um, and it should be the largest technical conference for AI.

• 1:00:58Jacob Effron?

  And ours is simple. Just, uh, we we we just run a humble podcast. So, uh, subscribe to unsupervised learning on YouTube. Well, thanks so much. This was, uh, this was awesome.

• 1:01:06Speaker 01

  Thanks for having me.

• 1:01:07Speaker 02

  Good to see you guys. Thanks for coming on.