YC Demo Day, Spoon = Bent (Again), iPhone Duo Reactions | Garry Tan, Om Buddhdev, Karen Serfaty, Antonio Li, Edward Ge, Alfredo Gonzalez

You're watching TVPN.

Today's Thursday, September 10th, 2026. We are live from the TVPN UltraDome, the temple of technology, the fortress of financial capital.

Capital.

Every time I see the road to Christmas, I laugh. How many days we got, Jordi?

105 days.

105 days till Christmas. Get ready. Uh, let me also tell you about ramp.com. They're making a musical or Broadway play. Uh, time is money. Save both. Easy to use corporate cards, bill pay, accounting, and a whole lot more. All All in one place. We'll get to the story, but awesome, awesome stunt. It also looks like it's a musical about Ramp Bill Pay specifically. I like that it's not just about the corporate entity. It's the subproduct. It's one SKU. You got to be making whole Broadway plays about single SKUs for your company. No, they got Billy Zane. They got a bunch of really talented people working on this. I think it's going to be really fun. The teaser trailer dropped on Eric Lyman's X account today, but It's gonna be fun to watch that project unfurl.

Very cool.

Well, should we do a little bit of a news roundup? We have Y Combinator Demo Day going on today. We got Gary Tan coming on the show, 11:25. Then we're going back to back with 5 different startups that are in this YC batch doing everything from humanoid robots to satellites to AI agents and credit cards, fintech. We're going all over the place, giving you a tour of what's going on in the startup world, which I personally enjoy because it's a good way to sort of ground yourself in what's happening in the— at the early stage. There's so much discussion about what's going on among the big labs and Apple, obviously, in the news yesterday. It can be all-encompassing. But the big story—

It's time to talk to the future.

Talk to the future?

The next generation? No, that's really what it gives you a little bit of insight into what things are gonna look like in a decade.

Yeah.

Some people don't think we're gonna be here in a decade.

They don't.

But that's okay.

Thoughtsense is one of them.

Never black pill. I saw a good post this morning. What'd you say? Josha Box said, if you think AI will lead to human extinction by 2030, I will take that bet. If we are still alive by December 31st, 2030, you pay me $1 million USD. If not, I will pay you $10 million USD.

That's a good, good, good joke.

Take it.

There's a lot of people out there that should take the bet, I think.

Why?

There's no— I mean, that's the whole joke.

No, I know.

Yeah, I mean, with a lot of these, the arguments break down. A lot of things break down. It's also just so difficult to put a PDoM over time, because you have to time-bound it, because, you know, General scientists will say humanity will go extinct eventually. The default number is 10 to the 100 years. That's a googol years from now. It's heat death of the universe. The stars collapse.

Yeah.

Even in a billion years, we're gonna be in trouble because the sun will be larger than the Earth, will have expanded to become encompassing the Earth. So you gotta get off Earth in the next billion years. We're working on that with SpaceX. Elon occasionally delayed. I think he's gonna make it before the next billion years. The AI thing, obviously much more pressing, all sorts of hacks and agentic capabilities, lots of debate going on, and a wall-to-wall press tour from a lot of the safety community. Daniel from AI 2027 was on Joe Rogan. It was an interesting listen. He makes a lot of good points. He breaks things down for a broader audience. And then Rogan will take him in these very funny directions, just being like, yeah, these AI agents are crazy. Let's talk about UFOs. It's actually like the meme.

Yeah.

They're like, oh, yeah, like, have you heard about remote viewing? And he breaks down remote viewing for him. And Daniel has to be like, well, I'm more sceptical of that, but I'm a real believer in x-risk. And so I think the net effect is There's going to be more awareness about the topic, more discussion about the topic. It might just wind up muddling it because people now put x-risk in the aliens, UFOs, remote viewing camp as like, oh, it's like a fun thing that Joe Rogan talks about. It becomes more abstract as opposed to being more concrete. But I don't really know. We'll see. We'll see. What did AI 2027 predict for September of 2026? They said this is the stage where AI becomes economically disruptive and starts accelerating AI research itself. We got the second part. The labs are all saying we have an automated AI research intern.

Just an intern.

Intern.

Just an intern.

Yeah, but that's accelerating and the models seem to be getting better at a faster clip. The economically disruptive is the capability. Jobs, junior software engineering faces major disruption. People managing and checking AI AI teams prosper. So junior software engineering, major disruption. We'll see. We'll see if that plays out this month. The job market has been pretty strong. We mapped that there's something like 17,000 monthly AI-related job losses, but that's less than 1% of the 1.7 million jobs that are lost every month. But then of course there are more than 1.7 million new jobs and new hires because people change jobs a lot. And so that's often what's going on, markets and politics. The scenario forecasts a 30% stock market rise during 2026 and a 10,000-person anti-AI protest and growing defense contracts. That seems—

yeah, it feels notable that we haven't— we haven't had a protest.

There's been a protest. There was a protest in San Francisco, a safety-led protest.

It was not nearly 10,000 people.

It was like more like under 100, but dozens. But yeah, I mean, it seems like the Joe Rogan experience, the national news tour, if you're going to get 10,000 people in the streets, this is a key step in the chain.

Well, Sheryl Crow was over on Instagram highlighting AI risk, and someone was saying, Sherry Oteri, is that a celebrity or something? Says, take to the streets like they did in Albania, all of us, so we can't be ignored. There's, I guess, calls for protests.

Yeah.

And there was another— there was another risk incident reported by The New York Times today. Anthropic says it blocked possible efforts to build biological weapons in a new report. The startup added that it could not determine if the research was legitimate or nefarious, leading the company to shut down the work. So some scientists were using an Anthropic model to work on bio— like biological research. And the Anthropic team flagged it as, hey, this could be in the in the line of a bioweapon. We don't know that that's what this person was trying to do, but we're going to shut it down because, of course, a lot of the research that goes into building a nuclear power plant is the same that goes into a nuclear bomb. And same thing, these are a lot of dual-use technologies. Anyway, anything else to say about Jacob Coxon who I think will have much more news? He went on Fox and Anderson Cooper, CNN. And I'm sure there will be a much deeper dive at some point. It feels like there's going to be a— it is interesting that with the launch of the Post and the Wall Street Journal article, there was not a 1-hour, 3-hour podcast. Like, we didn't get our situational awareness. Like, I want the 5-hour sit-down with somebody.

Based on the interviews, it doesn't really feel like he's got an hour of content.

OK.

Maybe he does.

But it's the best because you can just tell the story of the Terminator for an hour and take someone on that journey, and that can actually be in service of updating people about x-risk. You can tell the full story of how you think things might play out. There's so many different scenarios that are the bad ending worth avoiding. You can illustrate that. I don't know. Yeah.

Yeah, I thought it was—

I think it's coming. I think it's coming. Refresh your podcast player in a week. You're going to have the full podcast.

Let's pull up this clip.

I mean, he's got to go on Hard 4 or something.

Pull up this clip and let's watch.

Okay.

This is from Fox News last night.

I think that industry needs to be regulated as soon as possible.

Do you work with any third parties as far as this whistleblowing and coming out? Not at all.

This is entirely my own personal concerns about safety and things that I've seen personally and conversations I've had. He's serious that they think that industry—

a lot of people are keying on to the background here. They're like, wait, this guy just— he quit yesterday and he already has a podcast studio. People are like, is that someone else's podcast studio? And so I looked at the image, tried to figure out whose podcast studio he's in. It's Call Her Daddy.

Boom. Yeah, this goes way deeper than we thought.

It's fake Chiclets.

No, so I do expect it at some point as they dig in here based on how much sort of— it seems like there's very much a significant push from the overarching AI safety community to make this week all about this specific risk.

Yeah, he clarified this a little bit on Anderson Cooper saying that he worked with a bunch of friends. And where's the line between friends and professional? We're all a big family here. We're all friends. So, the answer could be muddled.

I don't know.

I'm sure we'll find out more soon. Anyway, you think he's going to launch a Neo Lab? I think he's going to drop a coin?

Book?

Movie?

There's going to be something. The story is not over. Story is not over. There's going to be something. Maybe he launches a Software as a service company.

It's crazy.

And then his spoon gets back. Yeah, um, trying to transition here.

No, it's, uh, somebody just said, uh, Sebastian says he's a D-cell with a crazy P-doom. That's a quote from Mountainhead.

Oh yeah.

Remember? Which is now out maybe a year, a year ago.

I remember you watched it.

Which is, um, yeah.

Yeah, but that was a sort of a niche film. It didn't, it didn't go as big as—

Yeah, I didn't think it was a good film, but it had some funny moments.

And it was about like an AI apocalypse, right? They go to like a— Yeah, yeah, yeah.

But it was around all about deepfakes.

They go to a bunkie.

Yeah.

Is there a bunkie?

There's a bunkie.

Okay.

Yeah. Maybe I gotta watch Mountainhead now.

But it's all about the doom scenario was around deepfakes.

Oh yeah.

Humanity thinking that there was like, you know, more conflicts happening, which would create actual wars.

See, that's great. I don't know that I'm super like worried about that particular scenario, but you could take me on that journey for 20 minutes in a long-winded answer. answer on a podcast, and I would be entertained.

Yeah.

And so, I think that there is material there. And I did see genuinely a lot of, like, sort of AI safety sceptics just asking for more storytelling around how things go bad. It's so easy to condense a PDoM down to a single number, but I think people like stories. They want something that actually illustrates the chain of events that lead to the bad scenario, as opposed to these vague things where it's like, I have a lot of authority, I have a lot of insight and 5% or 20%. That obviously, I mean, it does go viral, but clearly, there were people that had demand for more of a story that they could track and they could falsify individual. Keith Rabois yesterday was struggling with this where it was like, it's hard to falsify someone saying 10% in the next 10 years, but it's easier to falsify about like, well, the nuclear weapons will be fired at this time and this is what's gonna happen and this is how they get—

Also the job loss stuff.

Totally, totally, yeah. Anyway, let me tell you about CrowdStrike. Your business is AI, their business is securing it. CrowdStrike secures AI and stops breaches. Thank you. We have some of our team up in San Francisco for YC Demo Day. We're gonna be joined by Gary Tan in 10 minutes. We'll do a quick roundup of the rest of the news. Should we talk about Nvidia? Should we talk about Bending Spoons? What's interesting to you, Jordi?

Let's start with Bending Spoons. Bending Spoons acquires Miro for $1.79 billion.

Yes.

Paul Bonnet, a GP over at 20VC, says, but who gets the money? My usual breakdown. So this is 89% below the last round, and at less than 3 times revenue.

Whoa, Miro was valued at $17.5 billion?

Yeah.

That's really high.

Wow.

Yeah.

OK.

So anyways, we actually don't need to dive through all this stuff. Investors are getting—

Yes, but I am interested in the bending spoon strategy, because Airtable, Miro— Miro felt like modern— Collaborative slides and and and like mood boards, and then Airtable felt like the modern you know Excel sheet. And it feels like they're trying to maybe piece together a bundle, but I don't know if that's actually how Benning Spoons thinks about this. It feels like in the past they've been very interested in just it's a known brand with a strong business that they can run efficiently, and it stands alone. But for the first time, I'm seeing glimpses of. Something that could be like a portfolio or a roll-up where if you have Airtable, you get Miro for free or vice versa, and there's cross-selling options. Because if you like these more modern tools, like a Miro, like an Airtable, I've seen both of these deployed in the same startup. And the startup, there's always this tension between, wait, we're paying $20 a seat for this, $20 a seat for that, or we could just do $15 for Google and we get all of them. So, you actually wind up with the portfolio, and I don't know, maybe the sum is greater than its parts. What do you think?

Potentially, but yeah, it was very— it's just insane to me that this was a $17 billion company just a little bit ago. And ultimately, this is why I have a lot of fear around some of these later-stage or even later-stage rounds that are happening right now because they're clearly— there's certain teams that are fantastic, great categories, great— on the path to being great businesses, and yet they feel so— many of them feel so far from being on that path to being priced at $100 or $1 trillion or whatever the number is.

Mm-hmm.

And we're going to see this exact same thing play out right now where there's companies that are around this valuation now that 3 years from now will sell at a 90% discount.

Yeah. I'd be interested to see if— have the metrics at Miro changed in the age of AI? Because you could imagine that people, if they need a collaborative mood board or slide deck, they can vibe code their own tool. But it's still a little bit of a hassle to deploy that and maintain it, even if you can clone it. So the SaaSpocalypse was fully averted by the Salesforces of the world, and the revenue decline never showed up in the SaaS companies. And so the SaaSpocalypse was canceled. But was there a micro-SaaSpocalypse that did show up at this level, and that's why we're seeing these deals?

I don't know.

It's not public.

Yeah.

Do you think this is all just like a promotion for David Senra's new episode Luca Ferrari? Because David Seder says, perfect timing. I have an episode with Luca Ferrari coming out this Sunday. Is that great? He's just like, thank you.

Thank you for lining up this deal with you coming on my podcast.

Promise.

Now, Jason says, paying 3x ARR at $600 million of ARR, over 750 customers paying over $100,000 a year, cash flow positive.

Man.

But Jamie Quint says, multiple years of 5% ARR growth will do that.

Yeah, if you're not accelerating— I mean, the entire multiple is based on growth. So there's not going to be a significant acceleration.

You've got to be growing.

Yeah, it's true.

So in other acquisition news—

Oh, yeah, Tailwind.

We've got a card out, but Tailwind got acquired by Shopify. That's very cool. Remember, if you remember, I think Tailwind was accelerating AI software development but wasn't directly a beneficiary.

And didn't they get like sort of like, you know—

A bunch of people in the industry, I think, came together.

I think Google, right? Wasn't it Logan?

Yeah, Logan was a big part of it.

Logan was like, I am going to provide a grant to keep these guys going. I thought that was awesome. That was like Such a cool community moment, and interesting that they now like found a proper landing spot, which is cool. Yeah, very optimistic scenario.

We had a little bit of iPhone Duo coverage yesterday. Yeah, we didn't see the crease.

We had a key question in the chat from Sally Analysis. They asked, "What do you guys think about the crease?" I couldn't see it in any of those Apple renders. Apple seemed to get really good with the. lighting placement to not show up the crease. Uh, true, my accurate take, I don't care about this. That is not enough of a crease to matter at all.

Yeah.

Especially if this is a mode where you're, uh, using 2 apps, you don't care about the middle at all. I have seen guys on football days have 4 TVs next to each other with full bezels. Like, you can see the line going down the center of the TV and they just don't care. So like, this is not a big deal. Now, the interesting thing is like, will it be an issue if you're like watching a movie on this thing on a plane? I think maybe, but I think the crease will probably completely disappear when you're actually locked in on a film. But I think the bigger thing is the aspect ratio is kind of weird for modern content because we have films which are widescreen and YouTube videos which are 16:9, and those will show black bars at the top and the bottom, or you could zoom in and then you're cropping the sides. Not a great experience. Some of the demos, some of the demo video that Apple showed was super cool because it was shot in this like 4:3. So maybe you go back and you get some 4:3 content from like the '70s and '80s, and that displays really well 'cause you don't need to do the pan and scan thing anymore. But, and then the vertical content will also look like bars on the side, like you're on an iPad. So I wonder if that will be annoying. It'll probably melt into the background. I think people will still like it. I'm very interested in the mix. Like, what percent— like, will this be as popular as the iPhone Pro has been relative to the iPhone? Will this be 10% of overall phone sales from this cycle? Will it be 1% or 30%? It's the fact that it's different, the fact that it's a way to stand out. It's their most foldable phone ever.

Yeah, this was a comment from—

Thank you to the listener.

Prolific. Spotify, the Spotify comedian Dave, who commented yesterday. It's their most foldable iPhone yet. It is, and it really is. This was we should have we should have this should have come to us on the show yesterday. But thank you, Dave, for for throwing that in the comments.

So it has yeah, it has something fresh. If you're bored of the same phone every year, this is a way to inject some variation.

So so really, it's like when they introduced the iPhone Duo two. Which will—

Quattro?

I think they're going to have to multiply them and get the iPhone Quattro. Then they'll say, it's our most foldable iPhone yet. It has 4—

Until it's just a full accordion, like a fan that just fans out its screens.

Yeah, the fan, the iPhone fan. It's just like this—

Yeah, 20 different folds. What was funny is I'm going to put you on blast here a little bit as you come in and you're like, yeah, this just makes me want an iPhone 18 Pro. And you were laying it out and you were like, it's their best iPhone yet.

They—

it's just a great phone. They made it thinner, lighter, faster, better battery life. And I was like, this is your joke. You fell for your own joke. And it— but it is sort of like this. It could be for some people like an anti-halo phone. For some people, it's gonna be the halo where they're like, I wanted something different. I want something bigger. I want the best thing. I got that. For other people, they're gonna be like, oh, thank God I don't have to deal with the foldable thing. Just give me the cheaper, $2,000 iPhone Pro Max 18 Plus, all these crazy things. Catalog variety is valuable.

Blake Scholl jumped in and said, when Steve Jobs introduced the iPad, he went to great care to explain why such a product category should exist to begin with. Apple seems to have made a pretty great version of a foldable phone without saying why foldable phones deserve to exist. I think it would be somewhat uncomfortable for them to explain Why exactly you need a bigger screen. It's like, hey, we know some of you guys are spending hours and hours and hours a day— This is such a bummer.

OK, continue. Just finish your argument so I can rebut it.

No, why don't you jump in, John?

I just think that there is— there's cool stuff to do on your phone that is actually less brain-rotty when you're on a bigger screen.

Like games on your phone, like which you would from time to time.

I would never watch Lawrence of Arabia on this screen, but you give me a screen with twice as much real estate, I'm going Lawrence of Arabia. I'm going Citizen Kane.

Yeah.

No, and I agree that that is the use case.

Yeah.

But I think it's somewhat uncomfortable because of the overarching sort of discussion around phone addiction and screen time and all this stuff.

Yeah.

It's bad for kids. It's probably not great for adults. And for Apple to come out is like, you know, we know that you're spending hours and hours every day, you know, getting mindless entertainment from our devices. And so, we get— we're giving you a bigger screen.

I don't think Apple needs to touch that. I think that like the things that they promote on their phone are good. Like, you can answer your emails, you can connect with your friends, you can text them, you can watch educational videos. There's a positive flip to everything that's like a negative example. Yeah, there's some game that's brain rotty, but there's also some game that's really fun that you remember. You talk about Rust and Halo. Like, these are core memories for a lot of people. It's a positive thing.

Yeah.

So being able to do more of that. Plus, it's just like, you don't know when you're— it's very hard to plan ahead for like, oh, I'm going to get caught with an extra hour waiting at the DMV, or I'm going to wait on a plane, or I'm going to have some downtime, and I'm going to want to actually like enjoy something more thoroughly. If you have the Duo, it's your daily driver. It's always on you. You didn't need to think, oh, I should have brought an iPad because I'm going to be stuck waiting in this line for 20 minutes. So I want to watch something nicely on a really big screen. You got the Duo in your pocket. You're good to go.

I'm not sold.

Get me a sales job.

I'm not sold, but get this guy a job at the Apple Store.

Let me tell you about Codex. Codex is a powerful workspace for getting work done with AI agents, whether you're writing code, analyzing data, creating content, or automating business workflows. Codex helps you move projects forward from start to finish. We have Gary Tan in the waiting room. Let's bring him in. Let me even keep him waiting. I want to know, Gary, how you doing? And are you going with the Duo or the Pro Max Plus 18? Do you want a folding phone?

I'm down for the folding phone. Okay, that sounds great. I got the Samsung trifold. It's pretty sick.

You do? Whoa, so you're living in the future. Personally, I'm holding out for the iPhone Quattro. You want the fourfold screen? I wanted to Fold out like an accordion.

I would buy that as well. I would also buy that.

Also buy that. Okay. Well, uh, how, how are things going?

Wait, please. Like, yeah, it's top of mind for everyone.

Yeah.

YC Demo Day is also top of mind, but are there any new ways that you use the tri-screen phone?

Oh, interesting.

Outside of just more entertainment, I'm struggling to see.

When does it go full Full trifold for you?

Uh, you know, like just watching, watching TBPN mainly.

There we go.

There we go.

Okay.

Yeah.

I want to see your beautiful faces, and, you know, it's just full resolution, you know.

Got it.

Got it. Um, awesome. Well, great to have you here today on another absolute banger of a show.

Kick us off with the stats. Like, where is YC? How big is this batch? What differentiates this batch and the we'll go into the broader technology landscape, the startup ecosystem, all that.

That sounds great. Yeah. I mean, it's 200 companies presenting today. About a quarter of them are hard tech. That's up 40x since our low bar, maybe like 3 or 5 years ago. I mean, hard tech is in, it's big, and it's here because of AI. I mean, intelligence is on tap now, right? So what that means is instead of just doing Bits, like it's time to do atoms, guys. It's time to actually create abundance for one another. And then the kind of stuff that people are doing, completely insane. I mean, it's some pretty pure research type stuff. We got a company called Frontier. They made biochips that use basically neuronal cells.

I think we had them on the show already. Yeah, I think he went viral earlier and we got him on the show.

That was Brain in a Vat.

Yeah, we were doing Brain in a Vat week. We had like 3 companies that were doing bio stuff. It's a crazy time. On the hardware question, is it— are we seeing a boom in hardware startups because there are more cool things to do with AI? I'm thinking of— I have a sticker box. You put— the kid pushes a button, they say a prompt, and then it prints a little cartoon sticker. It's very cool, wouldn't exist before generative AI. But then there's the other side, which is like hardware has always been hard. You have to have these long cycles, you have to deal with manufacturers. And I feel like AI can speed that up by going and getting you quotes from a bunch of different factories, making sure your supply chain's in order, doing inventory management, all these different things. Is it both? Which one are you seeing be the most acceleratory?

Oh, I mean, look at what's happening with GPT-6, Navier-Stokes equations. That's insane, right?

Yeah.

So, so if you're trying to board layout, it's, you know, it's— you're going to be able to do it much faster.

Yeah.

And, you know, what, what was the alternative to that? Like, if you are a team of like 1 or 2 people in a garage, the chance of you actually springing the person who really understands some hard tech thing that you really need, I mean, it's so close to zero, you know, or you need to have raised $100 million to convince that person to quit. Now, you might not need that person on day 1. You could just go to GPT-6 Astra.

Yep.

You say, hey, look, this is what I'm trying to figure out, and you're iterating with it. You actually have a demo that's 10 times better than what you could have done without hiring that person out of that lab or, you know, that university. Like, you don't need to convince them.

Yeah.

You just actually get it farther along. You can show it to an investor, you can get money, you can get there on $5 million instead of $50 million. It's sort of like, you know, multiple stacking things that are happening. Like you have intelligence on tap, then any given founder who can identify a thing that people need in the world, suddenly you can say, you know what, this is the path. This is like, this is always what we've done at YC.

Yeah.

Is, you know, we're giving you half a million dollars. What can you actually get done? You know, these days it's half a million dollars plus more than $1 million in cloud credits, right?

Wow.

So what's $1 million in cloud credits if not like maybe the equivalent of $10 million worth of R&D for, You know, whatever.

I remember back in—

Caltech, Harvard, you know?

I remember back in 2012, it was $170K, $150K, and then $150K in cloud credits. And it was hard to spend those credits. Like, you had to actually get things to scale, because it was only if you had the demand side solved, then your server hosting could be deferred a little bit. But from actually going 0 to 1, that, that $150K on AWS was sort of just like, I wish I could use it. And people were joking about like, maybe we should mine Bitcoin with it, but you couldn't actually use it in the R&D phase. And now you actually can pull those credits into the R&D phase in the form of tokens, which is a very different dynamic. So I imagine that the credit usage rate has gone up a lot.

Yeah, people need to be token maxing. I mean, what's funny, I mean, yeah, you know, OpenAI will come and just give any YC company $2 million worth of tokens, right?

Yeah.

Yeah, that's great.

I mean, that's a part of the— that's a part of the deal now, right? Like, it's optional. You don't have to take it.

But yeah, yeah, yeah.

Uh, we tell people, look, like, if you're doing something hard, yeah, you should probably take the $2 million because, you know, if you can't figure out how to turn $2 million worth of tokens into $100 million worth of enterprise value, what are we doing here? Yeah, what are we doing here?

Yeah, yeah. Um, what do you— what are you seeing on the hard tech side in terms of distribution and go-to-market. One of YC's strengths has always been selling into the YC network. You start a payroll company, you can get the other batchmates to test it out, you get your early adopters there. And I noticed that this was happening a little bit in the Gundo, where you would see, uh, one guy's running a machine shop, one guy's running a cloud seeding, one guy's running a drone company, and they'll run across the street and say, I need a new drill or hammer or whatever they need. And But I'm wondering, in the, in the hardware sense, are there, are there YC companies that are actually able to get products in the hardware space that bring value to other startups? Or is it more of like camaraderie around we're in the hardware world, we'll communicate and share notes, but we're not actually selling to each other on day one?

I mean, you don't really have to necessarily even sell to each other, though. That's happening like the space ecosystems coming together. Like StarCloud is a great buyer of lots of other space tech. That's how that's been happening. I mean, we got defense tech coming together and people trading notes like we're putting together a cradle so that we can actually help all YC defense companies get in theater with real Department of War departments right off the bat. There's just a lot that we can do. Next week we actually have a, a D.C. defense tech conference in D.C. We're bringing in a bunch of our top defense companies and they're meeting the policymakers directly. They're meeting the people from the Pentagon directly in their own backwoods. And that's actually stuff that YC can do now that, you know, it just increases the chance and likelihood of people actually succeeding. I mean, defense tech is the wildest thing that continues to happen in this administration. It's that I think, you know, huge credit to the people in the current administration because they're realizing like, hey, we got a drone war happening. We have an AI war, and America needs to be— can't just pay, you know, cost-plus to these defense, you know, the sort of like industrial complex that exists today. Like, we actually need— and they're willing to give 7-figure contracts to teams of like 2 or 4 people who like suddenly become teams of 20 or 30 people plus like hundreds of agents.

Yeah, it's been remarkable how how many opportunities that have been there, even in the post-Palantir and Anduril world. There was a time when I think people thought that there might be more of a winner-take-all dynamic. It might be more like social. But there's been so many teams that have had impacts in various ways.

Well, it's been exciting. So 25% of the batch is hardware, up 40x. What about the other 75%? What are the trends that you're seeing on the software side, dev tools? I think a lot of, like, part of why some people are driven into hardware is they're looking at how easy it is to make software now and how fast it is and thinking, well, I don't want to go comp— if I can build this in 2 days, someone else can build it in 2 days. So maybe I want a hardware component. But at the same time, we're still seeing a bunch of early-stage software companies rip. What trends are you seeing across the board with agents and infrastructure and all those things?

Yeah, this is, I mean, honestly, this the golden age of software as well.

Mm-hmm.

And it's interesting because, I mean, my prediction would be you look at, uh, the public companies of today, uh, any company that's got thousands of engineers who are doing it the old way, like they're gonna be replaced by a YC startup or a company like that that has maybe 20 people working there. And that those 20 people each will do the work of like 100 engineers themselves, right? So you got like a 2,000-person workforce. And you're just like blowing them out of the water. And so that's going to be the biggest shift. Like you're going to have AI-native startup companies that are just brand new that have started in the last 2 or 3 years. They don't have any of the bad habits that were developed over the, you know, sort of ZIRP era. And they're just going to come in and out-execute every single SaaS company that doesn't do it that way. And it's good. It's going to be pretty epic. It's going to be a turnover of epic proportions and— the world of software. And, you know, that's sort of the rest of the story.

Are you seeing pitches for, uh, like NeoClouds? People that want to go through YC?

Yeah, yeah.

I mean, Wafer's killing it. Shapeform is killing it. I mean, uh, you know, the amount of inference in the world feels like it's going to go up like 10,000 to 100,000x. Like, people are just not really prepared for, you know, it's actually kind of the funniest thing. Like, you can work in SaaS where, uh, you know, things are going to grow 10x. And then or if you look at inference, you look at data centers, you look at semiconductors, you look at GPU and even CPU demand, it's going to go up like 1,000x, right? So that's even it's even bigger an opportunity. And I think people are just not it's not priced in. I you know personally I think is it's not priced in at all.

Where does are you still using the term little tech?

I always like oh definitely.

I mean yeah yeah yeah where where does little tech stand on the AI slowdown debate, regulatory capture risk. When I look at some of the companies that we're going to have on the show, I feel like they might not be affected positively or negatively by the discussions in Washington. But how are you seeing it broadly? Because typically regulatory capture can, you know, harden the position of incumbents, and that's bad for little tech. But How have you been dealing with the current conversation?

Yeah, I mean, I, you know, I think we should be talking less about this Jacob Coxon guy. We need to be talking a lot more about what is actually happening with Hugging Face. Sure. Like, are agent swarms going to come and take over infrastructure en masse? And then what are we actually doing about that? Like, I don't want to hear about some guy who worked at Anthropic for 2 months. And I don't really care that there's a, you know, there's a coordinated effort to try to influence politicians to sort of do a knee-jerk response. And like, the thing is, like, we shouldn't be— that's a smokescreen. You shouldn't be paying attention to that. We need to be paying attention to what is the actual thing that we can do to prevent, you know, agent swarms taking over entire data centers.

Yep.

Right. Like, what's our— what's our shutdown strategy? Right. Like, how do we ensure provenance? Where is this agent actually located? Like, you know, what software can we build and what cybersecurity defenses can we build today? And like, Like, you know, that's the level of discourse that I think we need. And we just don't have that, right? People are just saying, you know—

Well, discourse is one thing. Are there— is there a meaningful number of YC companies now that are actually focused on that set of problems? Because one way to think about AI safety is to go on podcasts and talk about it and write essays and write blog posts. And another way is to actually build cybersecurity products specifically with, like, you know, agent swarms in mind. And, you know, thinking of Ilya's post about NeoCloud Security, it's like, Well, I hope there's 5 companies that saw that post and were like, you know, I'm sure that Palo Alto Networks and CrowdStrike and all these companies are thinking about that opportunity as well. But it just feels like anywhere there's like fear, there's also an opportunity for commercial solutions to the problem that won't fix everything.

Yeah, I mean, we got that too, right? We got a ton of companies like SilverRail is able to do 100 milliseconds in in request loop prevent prompt injection.

Sure.

Right. So, you got an agent swarm. It wants to take on, you know, take over your AI systems. And like, it's in production now. Like, it basically prevents prompt injection and it can do it at a level that like, you know, it sort of doesn't matter how much smarter the underlying AI becomes and how much more powerful it is at cybersecurity. Like, if you have a kill switch on Entire requests because, hey, that's a prompt injection. Like, that's sort of the stuff that we need today. So I don't really care about science fiction, man. I saw Terminator 2 also. Like, that's, you know, we're not here to talk about that. Like, we need to actually talk about what's really happening with the servers, what's actually happening with the agent swarms, how do we actually prevent that? And I think that, you know, little tech has a role in actually solving— actually providing the software, the services, the hardware, the things that actually solve these things. And then when it comes to regulation, it's like, let's actually pass regulation that prevents the thing we actually care about instead of whatever a socialist says on the— in The New York Times. I don't care about that.

What, what are you telling YC startups not to do right now? I'm thinking there's opportunities in like selling data right now. If you just wanted to generate as much revenue as possible before Demo Day, maybe Brokering data could be a way to do that, but maybe it's not the best way to build a super enduring business. But maybe that's one example. But what kind of things are you—

I mean, I think cases like Mercor are very, very— they're body shops at some level. And then we're seeing this new wave of data companies like AfterQuery or DataCurve that basically beat them head to head. And they don't do it by being body shops. Like, you don't need thousands and thousands of people. You need, you know, dozens of people. I mean, it's sort of the same deal. Like, you look at, you know, a Workday, it's full of thousands of engineers who are not, you know, using agentic systems. And like, I would argue that like a lot of those, you know, first wave data companies don't— like, they're basically kind of like, you know, glorified labor marketplaces, right?

Mm-hmm.

But the new wave of data, like, we have more than 10 companies making more than $10 million a year in data. And then the top 2 are making, you know, hundreds of millions, like on the way to billions.

Wow.

And so, you know, there's a new wave, and those new wave companies do not have 1,000 employees and won't ever have thousands of employees. They will have 20 or 50, right? And how are they doing it? They're actually doing it in a proper agentic way. And so, you know, that's like, you know, same thing in software, same thing in hardware, same thing in data. Like, there's a new way to build these things. And, you know, either you're building it that way or you're doing it the old way. Like, you know, people can still raise money. They can still go to Sand Hill and get that crazy growth round and people are just going to do it the old way still. But the new— the new way is going to come and it's going to wipe out the old way.

Makes sense. Gaming. A lot of people have seen Astro demos from the last week. You know, people just building GTA 6 because they're impatient. They get— they get something that is who knows how many hours of play it'll actually get, but it's an open-world game that has multiplayer components, things like that. Do you think that we'll see a gaming boom in the next batch of just small teams that are like, hey, I can kind of build anything that I want, and I can build the kind of thing that would have taken hundreds of people and years historically, and I can build it in 3 months?

Yeah, I just funded a company called Summer Engine, for instance. It's growing really, really fast. It's just like Roblox, actually. But, uh, and, you know, I think that that's the way to think about it, right? Like, code gen is opening up all this stuff, and then actually, like, shouldn't people— you know, a lot of the value is actually going to be in, uh, your guild, your people network, like getting your friends to come on and play, knowing when they're playing, like, uh, you know, designing all of that is actually the ultimate value. And like, and then I think it's the same thing all over again, like If you look at, you know, a company like Roblox, like, I don't think that they're building around like this idea that you could vibe code anything like this. The future's already here. People just act like it's not, right? And, you know, anytime that happens, there's like infinity, just opportunity. And, you know, the startups are finding it. They're going to go for it. And, you know, that's the future. I'm psyched for that.

Yeah, I would almost disagree. I feel like Roblox is going to come out with an amazing We have tools. Yeah, no, but even seeing like Astra being like, hey, we can make this even better. But there's still plenty of opportunities for early-stage companies to just adopt these tools even more aggressively and come up with entire new things.

Yeah, Roblox is the one that's like on the edge of the network effect. But we can debate it more later. Thanks so much for hopping on.

Yeah, great to see you, Gary.

Another fantastic—

excited to, to meet a handful of these founders and keep it up.

Yeah, always great to see you guys.

Have a good one.

Cheers.

Goodbye. Let me tell you about Cisco. Critical infrastructure for the AI era unlocks seamless real-time experiences and new value with Cisco. What do you think of the iPhone Duo animation? Marques Brownlee shared it on X. Do you love this? What do you think? So when you open it, it fades as if it's looking through the glass. And when it closes, it shows you what it's about to show you. It could just show you the actual outside, but instead it animates so that it looks like you're looking through glass and it appears 3-dimensional. What do you think?

It's so incredible. I mean, really pushed personal computing into an entirely new dimension. I think it's beautiful. No, it's beautiful.

I think it's beautiful.

Well done. Well done to the team.

And I think this is what makes Apple so special is that the folding phones exist. Somehow no one thought of this. How did they create an environment, maintain an environment where you can still go and experiment and think of a little touch like that that makes something that's been around for a while? Gary has one, a folding phone still make it special. And now, you know, you got everyone talking about this. So good stuff. Anyway, our next guest is in the waiting room. Let's bring in Om from Olam Labs, the co-founder and CEO. How you doing, Om?

Hey, you got my name right. Congrats on the pronunciation. How do people— I used to watch your YouTube.

No way.

People say Om. They say Om a lot. Yeah, I used to watch your YouTube videos, like all like the company documentaries and founder stuff a couple of years ago. It was great.

Thank you, man. Well, thanks for hopping on the show.

What do you do?

He was 8 years old at that time, John.

I'm 8 years old. Introduce yourself and the company, and then we'll chat about it.

Of course. Yeah, so we do RL environments and training data. We work with labs to evaluate behavioral things in models, like what models lie more, who's more deceptive, who collaborates better, all those kinds of things. So typical RL environments are a lot more about You know, how well can you complete this coding task? How well can you do this, solve this math problem, that kind of thing? We're testing a lot more of like the real, like we're aiming to create like real-world environments, you know, companies talking in Slack to each other. You have a PM, you have a CEO, you're playing diplomacy, negotiating, and pretty much like grading all of those things that are more like qualitative and more difficult to verify, but also a lot more valuable if we can start to measure it well.

Yeah. So what does the, what's the first step in building the company? Like when you started this, was it like, I'm going to go build the environment and then I'm going to go to labs and say, hey, I have this thing, do you want to buy it? Or are you just going to them first and saying like, look, I'll build anything, I'm going to build— like, what is the flow? Because the demand is so clear and high, I imagine that it's working very well. But what direction do things flow right now?

It's a lot more of the former of what you said. So like the latter, like the latter, like going to a lab and saying, I'll build anything you want, that only works if you have like relationships and a lot of trust with them or you used to work there. The form— like, when you're selling data, the best mental model is that you are selling a capability to a lab. Like, you are selling them something to get their models better at. So if you have a benchmark or you have a dataset on like, here's how you can get better at coding, here's how you can get better at, I don't know, like collaboration, like just anything, then you can sell it to them. That's why all these companies, like, you know, like any data company, see them posting like their evals on Twitter and trying to get like to go viral.

Sure.

Because that's like proof to the labs that, hey, I can make great data and like package it well.

Yep.

So that's like, that's like why, so the answer is the former. Like everyone's trying to show that we can make great data and then use that to sell to labs.

So do you have evals? Do you have benchmarks? How much time do you spend thinking about benchmarks or building new benchmarks?

Yeah, so I mean, uh, there's 2 things. So, so like making great benchmarks and publishing them is great for just like the broader industry. And like, especially in our case, we're like, we're measuring a lot, a lot of things that relate to like safety and all like the OpenAI Hugging Face stuff going on, the things about like how agents actually behave in, in, in these like multi-agent environments. That's more our domain. We are publishing stuff. We've released evals on how models all compete relative to each other, how much they lie against each other, how well they collaborate. We have a bunch more coming pretty much. We've had some tailwinds in recent times on the kinds of things that people are looking for is pretty much exactly what we bet our entire company on 4 months ago. It is working out well in that regard.

What What, if any, are some historical comps for your business? Like, what was your business 20 years ago?

Oh, God. Like, I mean, it didn't exist 20 years ago.

Well, no, no, no, no, no, no. I know. That's what I'm asking. I'm trying to understand, like, was there companies that were helping Amazon in the way that you were helping, like the labs, for example, where it was like this helping them deliver core product quality increases that we're scaling this way? And part of why I'm curious is because I'm trying to understand what your business looks like in 5 years. Yeah, 5 years, 10 years, and how you think about that.

I think a comp that I would give is like all of like kind of like the internet infrastructure companies. Like everyone who is doing fiber and internet and like, and CPUs and RAM, I would say that is like the closest comparable thing to data companies nowadays. Like you're building the foundations, like all a data company really is, like Merck, all those guys, they just go out and buy, like capture stuff from the real world and then give it to a lab, which isn't, you know, like if you're buying CPUs, you're just constructing stuff in the real world and giving it to like a large cloud to then like service their customers. So I think that would be the closest comparable thing that I can think of. There has been a question about the viability of these businesses. But I will say is that I think the viability of these businesses would be the exact same as the viability of, let's say, Nvidia. I think for AGI, for ASI to come, there are only 2 bottlenecks right now. We have no bottleneck with architecture or compute or agents or anything. It's data and compute. Compute is the biggest bottleneck. But now we're going to have 10x More compute coming online in the next two years, so then compute even more becomes data. So because that that is like like if you're scaling law pilled and if you're pilled in all the ways that these AI labs are, then then data is the like like like actual bottleneck. Like like the the new DeepSeek model that came out last night, like the reason that's like a crazy like dominating model is because they spent all the time on amazing data. That's it. Like all they did they do any like like most of the time was on. curating amazing data. That's pretty much all they did. But yeah.

Are you compute constrained?

Yes. Like, I think—

How so? Like, what is the nature of it?

So I think the best data companies are research companies. So one thing that we do is like when we create like post-training tasks that we want to sell to a lab, we test them ourselves on GPUs by like post-training like open-source models.

Okay.

So for that to like source GPUs and do that, you know, as we grow and as we scale and if you want to do like even more and more. I'd say at our stage, it's like, it's very easy to source the amount of GPUs that we need. I think especially like I see a lot of other companies that I know like in this space or just doing research broadly, even NeoLabs, they're all compute constrained. I think that is where the labs have their biggest advantage. You know, they have the highest margin on every megawatt that they make. So they will continue to like gobble up more and more compute faster than anyone else. So I think that if I was to name a risk, that would be like the main risk for like our business and other This is like it.

Andrew Ross has a comp, LiDAR flyover companies selling to Google Earth.

Oh, yeah.

That's a good comp. The only thing is Google Earth, it always was sort of— I guess it is constantly like the intelligence of Google Earth is always scaling, right? Because you can always get more resolution. Higher resolution.

I was also thinking about— this is a weird example, but movie production studios that sell content to Netflix. Like, you want the Netflix library to keep growing, so you constantly have to be making new capabilities, new content for the network. I mean, there's also just like in the Amazon example, it's like the sellers. Like, there's a company that makes the knives that are sold on Amazon and makes the appliances and stuff.

Yeah, yeah.

So yeah, all a huge ecosystem. Fascinating stuff. Hugging Face, very scary. Are you optimistic? Are you a techno optimist? Do you think it's solvable or are you just spinning your wheels before, before the single AI?

Well, I am optimistic. I've been optimistic my entire life. But, but like, I will say it is getting scarier and scarier.

All right.

Okay.

We were clapping for optimism.

Oh, okay.

Okay.

The part that is like, like very scary is that like normal RL environment companies that are selling like coding data, even they're now having to do safety grading because like they're seeing like when you When you're having Claude do a, like a normal coding task. Yeah. They're trying to like reward hack and trying to escape the sandbox into like normal environments. So it is getting like, like everyone's becoming, whether they want it or not, a safety company. Like just because it's, it is becoming like a critical issue for everyone.

Yeah.

Um, and I think that is only going to increase and continue, not like kind of randomly go away and get solved.

Sure, sure. Yeah, that makes sense.

Well, thank you. Uh, what numbers from your Demo Day pitch can you share with us? You want to hit the gong?

Oh yeah.

Uh, I can't. There are some numbers I can't, uh, like—

How many employees you got?

Uh, it's only me and my co-founder.

There we go.

Yeah. Well, congrats.

But needless to say, you're ripping. That's kind of the vibe I'm getting.

Seems like you're doing well.

Yeah, yeah. I mean, most of our metrics are public if you go through like our stuff. Yeah, okay.

Uh, Yeah.

Well, have a great day. Thank you so much for hopping on.

Great to hang. Come back on soon.

We'll talk to you soon.

Cheers.

We have Karen Serfaty from Agentcard, the founder and CEO, hopping on the show next. First, let me tell you about MongoDB. What's the only thing faster than the AI market? Your business on MongoDB. Don't just build AI, own the data platform that powers it. And let's bring in Karen. From Agent Card, talking about giving AI agents credit cards. Why not? Amazing shirt.

What is— let's start with the shirt.

Let's start with the shirt.

Let's start with the shirt.

What's on the shirt?

That is a great question. So this is Braxter. It's an AI character that we came up with.

Okay.

You know, this, this YC events are like an ocean of black t-shirts with very small logos.

Yep.

And it was very important to distinguish ourselves. So yeah.

Okay, smart. Uh, tell us about the, the character. It looks like a, a real person of sorts.

Yes, so it's like a banker from the '80s who does all our marketing. If you've been to our website, you'll see that he's, he's there promoting Agent Card.

Yeah. Uh, what is the elevator pitch for why not just use a standard card? Why not give your personal agent just your personal card? You Do a chargeback if it messes up maybe, but why do you need a new card?

Actually, we think that you should be using your real card.

Okay.

Agent card is not a new card that you give to your agent. We actually help you use your Chase card or whatever it is that you have. Interesting. Yeah. So the agent can use it in a way that's safe.

Yeah.

Because I fully agree with you. I think that when you ask consumers, people actually prefer to use their real cards.

Yeah.

Because you get points, you get disputes.

Sure.

Like all the things that you hear about. So, that is actually what we do.

So then, at what point do you plug in to a company that wants to enable agented commerce, to an agent platform that wants to provide more secure payment interactions? Who are you selling to?

Exactly. So, we work with companies building personal assistants. So, I'm sure you've seen online companies like Instinct, Orchid. These are the type of companies that are trying to enable authentic payments for consumers. And what we do is basically we're the payments platform that they can use so their users can use their cards safely.

Interesting. And then do you charge a fee that's based on the transaction size, or is there just a SaaS fee associated with using the service?

Yeah. So, we charge a monthly fee to companies. They get access to our main product, which is called the Vault. Vault allows companies to store customers' cards. And we have a second product that's called the Purchase API, which is basically a single API that these agents can use to buy from any merchant online, which is a very hard problem to solve. Like when you, if you want your agent to be able to buy something, maybe you've tried it, like you go to Codex.

Yes.

And you say, hey, buy me a table from Amazon. And what the agent's going to do is open a browser.

Yeah.

Start navigating.

Yeah.

Which is very painful to watch.

Yeah, I got, I got banned from a website for doing this exact thing. It was like, you're a bot, you cannot. I was like, I'm just trying to buy an expensive shirt. Like, give me a break here. They lost my business for maybe ever, but definitely a few days while I was on the cooldown from using an agent. What do you expect e-commerce merchants to do differently Because when I— my experience was having an agentic interface to their e-commerce website was going to accelerate my purchasing. I was going to spend more, more confidently. I was going to become a more regular customer. It felt very accretive for them. And yet clearly their current defenses and their posture was, we don't want that business.

Yeah. In my experience, merchants do want agents to be able to buy. They're just not sure about like how to implement that in a way that it's not a risk for them, which makes sense.

Yeah.

We think that the way to enable agentic payments and agentic purchases is to solve all this flow. So the first part is like, how do we give agents a card that they can use? Second part is how we help agents navigate these websites.

Yeah.

The third part, which is actually the hardest part, is is how do we help agents complete a checkout page? Yeah, because like the problem that we see now is that like we've been working on this, like Cloudflare has been working for ages now on preventing bots from completing checkout pages and now we need agents to be able to do it. So that is the, that is the problem that we solve. It's not only the card, it's the whole thing. Like how do we help merchants sell to agents?

How much of that is a technology problem? Because I, anytime I see A CAPTCHA, I think, okay, you could throw a lot of compute at this and beat it, or you could just meet with the CEO of that company and say, hey, let's, let's figure this out. We're going to do a deal that's mutually beneficial. How much of it is on the phone and on meetings versus technology?

I think that a big part of a company like this is the partnerships. But what I think too is that we cannot wait for the merchants to be ready. There are a lot of companies, this generation of companies that's building like this next consumer app. They need this right now and they are educating consumers so they can buy things with agents. So we need to solve this right now. Like in the meantime, we need to be talking to the merchants and we need to be talking to the PSPs and to everybody that's involved so they can enable payments for agents. But in the meantime, we need to solve it.

How are you, how are you sort of tracking the growth of of this market, it feels like it feels really primed to be something that 100x's over the next year if it's done, which means it's a great category to be building in as a startup. But how are you tracking it today? What are you looking at? Is it the instincts of the world driving most actual volume of payments right now? I know that Muse seems very interested in actually driving a bunch of conversions and, and purchases with their agent. But how are you looking at it?

Yeah, I think it's impossible to tell. I think that we have 2 intuitions. The first one is that we've been using agents to buy things for months now, and we know it's only a matter of time until someone figures out what's the right interface so consumers can do that too. I buy things from my terminal. I've been buying flights from my terminal with my agent for months. But I know that my mother is not going to do the same. Like, she needs something that's better. And I think that that's where companies like Instant come in, right? They are building on top of iMessage, which is like something, a platform that's a lot more familiar to consumers. We have no idea the size of this market. Our guess is that this is going to be larger than when e-commerce started because agents are faster. They can make more payments. They're more efficient, and it's just in my experience, it's just a better experience to buy with agents than to buy from a website.

Thank you so much for coming on the show.

Yeah, great to meet you.

Congrats on a great day. We'll talk to you soon.

Thank you.

Have a good one.

Goodbye. Talk soon.

Let me tell you about the New York Stock Exchange. Want to change the world? Raise capital at the New York Stock Exchange. Maybe you start at YC. Maybe you make it to the NYSE. Up next, we have Antonio Lee from from Nori Robotics. Not going to spoil it too much, but building humanoid robots, and the price will shock you. Shockingly cheap humanoid robots. Too cheap? Too good to be true? Let's bring him in and ask him. Antonio, we're going to bring him in in just a second. Tennis matches are finishing at 3:33 AM. We'll get to that story later. For now, we're staying with humanoid robots because we are joined by Antonio Lee, From Nori Robots.

Whoa, there he is!

Now I understand how it's so cheap. I can see the cost reduction. How much does that thing cost?

I want one.

Uh, $16.88 under 2K.

$16.88? $16.88? No, cheaper than an iPhone Duo. Well, cheaper than an iPhone Duo. That's crazy.

Hogging turnus.

And it has not one hinge but multiple hinges. How'd you build this thing? What's the idea? What is it actually capable Why should people buy one for $1,688?

Great question. So we built it entirely in San Francisco. We just basically put a bunch of motors together. Majority of it is 3D-printed plastic, and then we have an aluminum base below it. It's able to do almost anything. Well, it's obviously— it's a humanoid robot, and you should buy it if you want to do development on American humanoids.

Okay, so shotgun bench, dev kits, cannon shotgun That ramp cold brew on the desk. No, I mean, do anything within the API. I imagine you're not doing Boston Dynamics-level demos, juggling and dancing yet. But it's an opportunity for someone to come in and solve a particular problem. Do you see this as more consumer-focused or industrial? What do you— are there any forks in the road that you're designing around? Obviously, price is a really key differentiator. but where do you see your design decisions today leading to in the future?

Yeah, so we see it being used within the home mainly for now. Right now it's able to handle 80 to 90% of the majority of home tasks. But in the future, we obviously want to be able to deploy into hotels, grocery stores, nursing homes, anywhere that can be of help with the sort of labor shortage that we're seeing right now.

Sure.

Yeah, talk about the importance of just low costs in the sort of takeoff of humanoids. Because I think one of the— like, I know so many people that would buy an iPhone-priced humanoid just to mess around with it and have some fun. Honestly, I think a lot of venture capitalists will buy it just purely because they want to understand the category and the progress being made. But a lot of people are— when you're starting to see a humanoid that's priced more like a motorcycle or a car, it just gets into a different Like, people are gonna have such high expectations where they're like, hey, I spent, you know, I could have got a Model 3.

The robot dog, you can actually buy that, but it's like $50,000, and I still haven't really seen that many people pick them up for fun or for experimentation because they're too expensive.

Yeah.

Yeah.

Please.

Yeah.

So obviously, I think, like, when it comes down to early adopters, you need to have an initial low cost as a wedge into the market. Right now, humanoids are a little bit stupid. It's because the AI behind it is not very well developed. So you need to sort of have it more be like a larger toy in some ways and be within that price point to be able to appeal to a larger consumer base.

What's the weight limit for what it can pick up? We got a question from the chat.

It's around 1.5 kilograms per arm.

Per arm. Okay. So yeah, a plate, a cup, a beer, a Four Loko. It can, it can maybe get you a beer.

With the Monster Energy. What tasks do you guys use them for around the office and how do you actually control— what are the best ways to control it as a consumer? Are you just giving it verbal commands?

Yeah.

Asking nicely, of course.

So some of the easier tasks, like we have it grab us beers towards the end of the day and also grab us waters. So some of the simpler stuff right now, the majority of the controls come directly from LLMs so far. So we've seen actually an incredible improvement in the ability of LLMs to control robots at a lower level. So ChatGPT-5 Astro has been absolutely amazing for us, and it's been able to one-shot a lot of the tasks really well.

Yeah. How important is the development of agentic coding systems and these harnesses? Because it feels like a decade ago, people would grab a Raspberry Pi, but they'd have to write all this code. They'd spend days just wiring up some gadget that would turn on their lights It's around Christmas or something like that. This feels like even if you're interfacing with the robot at a pretty low level via an API, there's a lot more flexibility on the dev side to accelerate these demos and these experiments.

Yeah, of course. I would say the coding agents have been so incredibly important for robotics because it actually allows you to use a generalist robot in a generalized way. Obviously, if you take 10 hours for it to just code it to grab a beer, then it's not going to be very useful in the house. Whereas if it can take under like a second for the robot to write 1,000 lines of code and deploy to the robot, it's kind of a game changer in how useful the robot can be.

Yeah. So are you seeing— like, I can imagine I get this piece of hardware and I want to write a script that has it go from my living room to the fridge in the kitchen. And I can use an agent to write that code, but I could also be feeding the actual data live via the API to the actual model? Are people doing that? Are you successful there, or is it too slow at this point? Like, at what point do you stop actually writing any glue code between the model and the robot?

So I think in the near future, you still need a lot of that glue code. The main point is to make it as easy as possible for the LLM to use the robot. So for example, let's say if you want to move the robot from the kitchen to your bedroom, you can simply call on the onboard SLAM, which is a 2D LiDAR that has a navigation stack. And that makes it— it's like 2 lines of code. It just goes from one place to the other very easily. It's all about simplifying that and abstracting it away for the LLM.

Got it. What is your capability? You've built one. Can you build 10 this year? Can you build 100? I imagine there's going to be a manufacturing facility. There's a lot of different steps to actually scale up to any meaningful volume. But what does that road look like for you?

So right now we're making around 2 to 3 per day and we're mainly deploying to our early sort of research partnerships. We are manufacturing them all entirely in San Francisco, right down 10 minutes from here, from Tons of Points. And our goal is to basically by the end of 6 months from now to ramp up to around 400 robots per month manufactured here in San Francisco.

Yeah, that's not bad. So already doing 1,000, 1,000 a year. Next year, 10,000 at least, something like that.

What do you—

so, do you— how do you think that, like, where will the intelligence, like, specifically with the long term? Like, I can imagine, like, somebody just has, like, a Codex or Claude, and Claude just has fully access to the robot. And so you as the manufacturer, Nori, don't even necessarily need to provide the intelligence layer as long as the robot has the right sort of control mechanisms? Is that one potential scenario? Because I can also see the scenario where you guys end up building effectively your own end-to-end agent and you have local models and cloud models that help people accomplish tasks.

I think the way that we see it is it comes down to how fast LLMs progress in the next 1 year or so, and they have been progressing very, very fast for the past year. I think somehow, I'm not sure how they did it at OpenAI, they managed to give Astra the ability to understand 3D spaces very well, which has been a game changer for robotics control. This was thought to be impossible even a year ago. I think for us it ultimately comes down to, I think like we don't quite have control over how fast these big models or these big labs progress. So we're just going to be using what's on the market to give the best use cases and the best consumer product to our users.

Last question from the chat and we'll let you get out of here. Do these robots have lifelike hands? The hands seem a little shy below the table. What do we got? What are we working with?

Is it a claw?

It's a claw.

Okay.

So we—

obviously the cost is a pretty large consideration for us. We tested out a bit more dexterous hands. And from the results that we've seen, like you're able to do 90% of what a hand can do with a claw for now, just because the control policies are not quite there to fully leverage the dexterous nature of a hand like this.

Yeah, no, that makes a ton of sense. Uh, trade-offs and everything, and I think you made the right one. Well, congrats and thank you so much.

Very cool.

We have a, we have a drum set here at the office. We got to get Inori to just play the drums so when one of us has a good line, he can go—

that'll be good. Um, have a great rest of your day.

Great to meet you.

We'll talk to you soon.

Good to meet you guys.

Great stuff.

Cheers.

Let me tell you about Console.com. Console builds AI agents that automate 70% of IT HR and finance support, giving employees instant resolution for access requests and password resets. Um, we're gonna get to the tennis story, but we are joined by Edward Gee from Exosat, founder and CEO, building sovereign satellite infrastructure to compete with Starlink, going up against the easiest man to compete with, Elon Musk. Uh, what is the plan?

Edward doesn't look worried at all.

Not worried. What is the plan? How many satellites do you need to launch? What cadence do you want to get these up? What is the key pitch for Exosat to differentiate you in a very competitive market?

I mean, we filed for 11,000 satellites. We actually have a published ITU API and Shared Spectrum filing for 11,000 satellites. I think it would be the biggest satellite network outside of the United States, possibly even China too. The idea is this, right? A lot of countries don't want to rely on American defense contractors or Chinese state corporations for critical network infrastructure. And we are coming in and we're offering them something in the middle that's cost competitive, that can be launched on any launch option in the world, be it an American rocket, an Indian rocket, a Chinese rocket. that just ensures no single nation can block them from getting into orbit. And yeah, going up against Elon Musk is a huge proposition. But think about it this way. If we are even a mild annoyance to a $1, $2 trillion company, how much would we be worth?

Yeah, sure. You're in San Francisco, probably a Delaware C-Corp. Feels like an American company. How are you actually pitching this?

We are not a Delaware C-Corp. We are a Singaporean parent company. We're one of the only Singapore— So I think every YC batch has around like 2 or 3 weird companies, like for— they're either doing crypto or some weird reason like us where they are not a Delaware C-Corp. And we are one of these companies. So all of our engineering is done outside of the US. We know we don't use US vendors, we don't use US supply chains due to ITAR. And that means we can launch, build satellites that can be exported to and launched from any nation in the world.

Interesting. Interesting. The— are you basically going after like the long tail of countries? You mentioned it could be launched from a Chinese rocket or an American rocket, but it seems like you'll be selling to like India, Japan, loosely Western countries. But is everyone—

Lots of Global South countries like Kazakhstan.

Oh, interesting.

Nations in Africa, Indonesia, and Malaysia. We have huge, huge interest there.

Sure.

A lot of it is we found there's a lot of countries out there that want independence from US supply chains, want independence from Chinese government-owned infrastructure. And that's what we're really going after. The world is getting more multipolar. We're starting to see like a lot of countries start to essentially become like— what was the word again when it comes to supply chain independence?

Sovereignty.

Sovereignty. Sovereignty. Yeah, but sovereignty goes both ways, right? Sovereignty affects both moving away from both the Chinese sphere but also the US sphere. And we are the company in the middle that's giving them the salads that they want.

How, how do you approach launch? How do you get these things up?

Well, we've— we work with Indian, Chinese, and American launch providers. In the US, right, we can use like SpaceX, we can use Stoke Space, Nova Rockets. In China, we have the Landspace CQ-3 as well as the Long March 10B. In India, we have the Vikram-1 and the Vikram-2 from Skyroo and the PSLV launch vehicle. And in New Zealand as well, we have the Rocket Lab vehicles. So we can launch anywhere.

Interesting. What was your background?

So I'm a second-time founder building the second space company. Before this, I dropped out of school in 2022 to build my first space company called Aethero. At Aethero, we were doing radiation-hardened satellite edge computers and orbital data centers. And I scaled— we grew the company quite a bit. We actually got to millions of dollars in revenue and we actually put 2 satellites into orbit successfully. But the big thing was we didn't I think the biggest— so 2 things. I'm a huge believer in sovereign technology, right? So I think a future where just SpaceX and China dominate all of humanity's space infrastructure is not a good future to be in. It's like AI. You don't want just OpenAI, Anthropic to rule all of humanity's AI. That's a pretty bad outcome. So I think the other 80% of humanity should have a say in that future. And the second thing also is I think the biggest space companies in the US and China have already been founded, right? I think there's so much competition going after after these big anchor customers like the US military or the PLA in Chinese variant. Uh, but like, space is a field where incumbent advantage matters so much, and that the best move to do, uh, might be just going after everyone else, going after all of these countries that traditionally haven't been prioritized by space companies. And that's what we're doing.

Amazing. Well, congratulations, and thanks for coming on and breaking it down.

Yeah, very cool. Great to meet you.

Have a great rest of your demo day. Goodbye.

Cheers.

Let me tell you about Shopify. Shopify is the commerce platform that grows with your business and lets you sell in seconds online, in-store, on mobile, on social, on marketplaces, and now with AI agents. And I'm also going to tell you about Railway. Railway is the all-in-one intelligent cloud provider. Use your favorite agent to deploy web apps, servers, databases, and more while Railway automatically takes care of scaling. Monitoring and security. We have our next guest, Alfredo Gonzalez from Momanta, coming on in just a minute. Just a minute. There are so many good iPhone Duo reactions. The toughest thing for Apple's marketing team, the sequel. They're going to call it the iPhone— Apple Duo 2. Is it not the iPhone Duo? It is the iPhone Duo, right? It's not the Apple Duo.

No.

It's not a 7.

iPhone Duo.

iPhone Duo 2, iPhone Duo 3. That's going to be— it's going to be a mouthful. They never did— they don't have iPhone Air 2 yet, do they? It's just iPhone Air and then they just stopped it. They haven't refreshed it yet. Air Max. iPhone Air Max 95. They'll get there in 94 years, something like that. Well, anyway, we have our next guest. Let's do it. Let's bring in Alfredo from Omonta, building a personal research lab around individual patients.

Alfredo, how are you doing?

I like where this is going. I like the one-liner.

Thank you so much.

Thank you for having us.

Thanks for hopping on the show. Congrats on a great Demo Day. Introduce yourself and the company.

Yeah, so a little bit about me. We're— our team is essentially biomedical researchers that have been dedicated to just improving patients' health outcomes. for the last decade. And we realized that there's a huge gap right now with the resources that patients and clinicians have as far as being able to go really deep into their data and ultimately understand the biology of their tumor and their diseases. And that's what Omonta does. We're laying down the infrastructure to be able to essentially have a whole team of LLM and AI agents that are supporting patients and clinicians with their care, helping decoding their specific disease, their biology, and finding not just the ideal therapeutic path but additional therapies that would have been missed otherwise. So that's kind of our—

So you scale through doctors? Like, if I have a tumor, I go to a doctor, they biopsy it, send it off to a lab, there's some raw data. The doctor might be interpreting that for me, having a conversation for me, but they might also give me a report Do I get a report from you or access to an app from you? How do I see the results here?

Fantastic question. I appreciate you did your homework there. That's definitely much more knowledge than most people have. But, yeah, so as of right now, the way that the medical system works is they do take your sample and they do these very limited sequencing panels, typically about 50 to 500 genes. Our whole gene space is about 20,000+ genes. So, it's trying to debug and not having access to the whole code base, right? And to be able to fully do this, it's knowledge-intensive work. But now, in the era of LLMs and protein models, that's where instead of having one single person that could do all this, you have a whole team of agents that can do that deep dive.

So, there are labs out there that can do the full sequence. You might advise a doctor to pick one of those labs because it's not just going to be more data, it's going to be more data that's actually actionable thanks to your platform.

Yeah. So, our platform is that missing link that helps the patients, the clinicians go deeper. Because as of right now, what they have is these very static reports.

Sure.

Right? So, they're not dynamic enough to be able to keep up with the pace of discovery generally or do discovery in real time for the patients. Yeah.

Do you have a pretty typical business model like insurance reimbursements or like the clinic will pay for this as a service or is there something else going on?

Great question. So, right now, it's direct-to-patient, direct-to-consumer, or direct to a sponsor. Ultimately, though, you know, there has been so much research showing out that the earlier you do would do the sequencing and the informatics, you actually can significantly improve the median life expectancy of patients but also reduce the cost on insurance payers. So, it's definitely something where we're already seeing the signs of insurance starting to cover— actual insurance covers sequencing for recurrences and advanced tumors in most developed countries, right? And you are starting to see already how insurance— the tailwinds are there. Insurance will eventually cover this because ultimately it is— Cheaper for the payer insurance, and you're also improving life expectancy.

So, yeah. What is the biggest bottleneck right now to AI progress in cancer detection in medicine?

Honestly, I think it's the infrastructure. So, it's a fairly huge data science problem, right? I think we have these legacy systems that are built around megabytes of patient data. outside of imaging. But the genomic sequencing, you're talking about hundreds, like 200 gigabytes, 30 to 200 gigabytes. It's a significant data science problem. So the infrastructure's not set up yet for these systems to fully take advantage of it. And that's in a sense what we're also setting up.

Very cool. What kind of traction are you sharing at Demo Day?

Oh, so we have launched almost a month now ago, and we are at $319K MRR with over 150 patients that have—

Crazy.

Bang.

Well, congratulations.

You got to lead with the ARR number. Pick the bigger number. I thought you were going to say $319 ARR.

You had me in the first half. It was good. Well, congratulations. I'm sure it'll be a successful demo day.

Thank you so much. Yeah, and it's good to be able to help patients and their families and help improve the health outcomes for everyone. So thank you so much for the support.

That's exciting. Great to meet you, Alfredo.

Have a great rest of your day.

You bet.

We'll talk soon.

Talk to you soon. Goodbye. Let me tell you about Figma. Agents meet the canvas. Your AI agents can now create and modify your Figma files with design system context. Okay. We got through demo day. We got to go over to tennis, tennis, tennis, tennis. There was a match that ended at 3:30 a.m. and people are not happy about it. The Wall Street Journal's headline is, no, no, no, no, no, no, no, no. There's nothing cool about a tennis match that finishes at 3:33 a.m. It's ridiculous. Sorry. But tennis at 3:33 AM isn't cool. It's ridiculous. What happened? Bill Ackman loved it. He said, one of the greatest tennis matches ever, and very few got to see it live. This needs to be fixed now. It's unfair to the players, fans, and sponsors in the sports. What is the actual proposal? What do they want to change? Start the game earlier or cut it off entirely if it keeps going?

Yeah. So I'm trying to figure out why this ever was, uh, is or was a thing. I, uh, hopefully the Journal article can, can break it down, but I watched a little bit of this game.

You did?

I went to bed. It was, uh, yeah, I think I went, went to bed. It was actually a later night. I went to bed around 11, um, but the game was going at well past 3.

Yeah.

Ben Shelton, who actually won the match, it was an upset, uh, was hitting 100 146-mile-an-hour serves at Carlos. Carlos at one point was shooing off the camera crew because they were filming him and he was actively going to throw up. He was pushing it so hard that he was going to throw up into a towel.

Yeah.

Really, really crazy. Anyways, Wall Street Journal says sometimes this glorious sport Can't get out of its own way. Everyone is expected to pretend night tennis is a fabulous experience. We're supposed to deploy the word epic and act like we didn't brush our teeth and hit the hay after Shelton took the second set. I do actually have a family member that did exactly that, like blowout second set. They were like, all right, like, we're good, we're good. And it ended up being this incredible game.

Okay.

The US Open has long cultivated the idea that the wee hours are part of its identity as much of— as much a part of the formula as the $23 honey deuce cocktails and the plume of cannabis smoke wafting in from a nearby park. Even ESPN feigns enthusiasm for the après-midnight when it would never schedule a tentpole like the Iron Bowl for such an uncommon hour. It's a city that never sleeps. Why should tennis? Okay. How many 3:30 AM finishes have you attended at Yankee Stadium? I appreciate the romance. New York remains a restless town where you can get a steak dinner and a tattoo on your back long after quitting time. The subway is 24/7 theater. You can crawl out of a bar at dawn. There are amateur bike races that start in the dark at 5:45 AM. It's zany, but Alcaraz versus Shelton isn't a corned beef sandwich at an all-night diner. This sounds like AI slop. It's a marquee showcase for sports.

No, I've been reading this and I think I think it's just like fun turns and phrases.

No, it's good. It's a marquee showcase for a sport that needs every bit of US attention it can get, and it got buried like a Tom Snyder sit-down with Buddy Hackett. Even people in Hawaii were like, this Alcaraz-Shelton match is running pretty late. This isn't a routine to celebrate. Okay, I'm sorry. A lot of, a lot of, a lot of, we're gonna have to run this through Pangram.

Parallelism.

Yeah, this isn't a routine to celebrate. This is when someone needs to step in and scream, what are we doing?

You know, I found 2 new AI phrases. I don't remember both of them. One of them is, in our increasingly online world, when someone's writing about online culture or something that happened on the internet, or like with the boom of social media, like there's 2 phrases that were popularized Because journalists who needed to write about things that happened online kept using this phrase, that got baked into the LLM, and then now it's getting spit back at people. And so a lot of people are turning away from the phrase in our increasingly online world. So avoid that one too, I guess.

Sash Kane says nighttime, you wouldn't understand that. It's a New York thing.

Yeah.

Yeah, it's funny because I feel like you can maintain the the mystique of the US Open by just having games wrap up around midnight. Like, it's still very much nighttime.

Totally.

Like, it's dark out, you know, people are sleeping. Anyways, definitely seems just incredibly rough and unfair to the players. This has happened in UFC a few times where the— in order to hit a US time zone for the pay-per-view, the actual fighters have to be fighting at the equivalent of like 6 AM.

Oh, yeah. Where they are.

Yeah.

Which is just— imagine waking— imagine one going to sleep at like 10 knowing that you have to wake up and get in a death match at 6 AM. Sounds pretty rough. The alternative is just to stay up and like, you know, just try to be good.

Because it starts at 6. So, You got to get warmed up.

Like, the whole event is starting at probably midnight.

Well, I guess the move would probably be just don't adjust for jet lag, like get there on your schedule and just treat it like it's a 6:00 PM fight for whatever your local time zone is, I suppose.

I don't know.

What do you think about this? Chairman Bernanke, you can eat the same thing for breakfast every single day for like 20 years in a row and no one will say anything. It's normal. You eat the same thing for dinner for a month straight and people call you psychotic. I think it's fine to eat the same dinner every night for a month.

What's wrong with being a psycho? I used to do chicken and rice every night for months, for years.

Yeah, that's a good point. Let me tell you about public.com. Investing for those that take it seriously. They got stocks, options, bonds, crypto, treasuries with great customer service. And they got more. Anything else in the timeline before we get out of here? We do have to cut the show a little bit short today. We have a little bit of family stuff going on, but we will be back tomorrow at 11 a.m. sharp. We have a great show. Alex Heath is joining us. He has some big news. He's becoming a venture capitalist, putting some money on the line, finally risking it all in the markets every day.

Putting putting down his capital J journalist hat, hanging it up.

No, he's keeping the hat.

He's keeping the hat.

He's keeping the hat. He's keeping the capital J journalist hat, but he's also going to be ripping some checks.

Okay. Ripping taking some taking some flyers.

And yeah, we got Saiger and Jetty coming on tomorrow. I'm sure we'll talk about that AI safety stuff because he's been a friend who's been evolving.

And hopefully we get an update on Saigerbet features.

Definitely. I've been enjoying seeing the Sauger bet references in the chat on and off. But leave us 5 stars on Apple Podcasts and Spotify. Sign up for our newsletter at tbpn.com, and we will see you tomorrow at 11 AM Pacific sharp.

We love you.

Goodbye. Have a great rest of your day.

Have the best Thursday afternoon of your entire life.

Goodbye.