Darshan Shanker, Pavel and Delian Asparouhov, Alex Konrad, Jordan Schneider, Willem Van Lancker, Christian Garrett, @carrynointerest, 11x Controversy, Robo Arms

You're watching TVPN. It is Tuesday, March 25, 2025. I got it right this time. We are live from the temple of technology, the fortress of finance, the capital of capital. Show starts now. We got a great show for you day folks. We got a bunch of call ins. We got some breaking news in the financing world. Companies are doing deals.

We got deals are getting done.

We got, we got media announcements, we got VR announcements, we got news about China coming on. We got a bunch of people calling in. But we also got some news. I see you're enjoying some Lucy cinnamon.

It's hot today here in Los Angeles. It's feeling like spring, John.

It's great. Well, what's your take on Signal? Because the Trump administration apparently I've heard all these defense tech founders complain, oh, I got to use Microsoft Teams because my work is so important and like I'm saving the country. But it's such a bummer. I wish I could use slack. Turns out they could just use Signal.

I mean, they all use Signal. I guess it's very common.

Maybe you should just run the entire organization on Signal. Seems like you just do it.

The US Government on signal and ramp.

It seems like it already is running on Signal. That's of course, the news of the day.

It's just wild that I try not to follow any political news. I try to keep my head in the sand on that kind of stuff. Yes, but I did see their emoji usage.

Oh, really? I didn't actually look at the story or anything. I saw this as a tech story and that's the way we're changing treating it. Trump administration was found using Signal. Atlantic's top editor said he was added to a text group in which top US Officials discussed detailed plans to bomb Houthi targets in Yemen and with other U.S. officials. An extraordinary breach of security from an administration. These were the repeatedly vowed to clamp down on leaks. Oh, that's why that's going viral. Yes, the. It's the fist emoji followed by the American flag emoji followed by the fire emoji. Very American.

Very American.

Not.

Not super tasteful.

Atlantic situation Editor in Chief Jeffrey Goldberg recounted in a 3,500 word story published Monday how he got a connection request on the Signal messaging app from someone identified as Michael Waltz. He initially believed the request was fake, but Ray later realized the account belonging to the US National Security Advisor was genuine after the group discussed detailed plans for an attack on the H. A militant group that has carried out numerous attacks on commercial vessels in the Red Sea. Goldberg didn't publish the actual plans in the article, but he said Defense tech. Defense Secretary Pete Hegseth at one point shared a post that featured operational details of forthcoming strikes in Yemen, including information about targets, weapons the US Would be deploying, an attack sequencing. Hours later, the attack went ahead. So it's kind of, I mean, there's a bunch of, like, bad takes about this one is like, oh, Signal got hacked. That's not what happened. They literally just accidentally added the wrong person to the group. Signal was not compromised. The encryption still works. The other kind of mediocre take or bad take was like, the Atlantic shouldn't have posted this.

It seems like the best possible marketing ever for Signal because it's very quick, easy to understand. Okay, this wasn't a hack.

Yeah, it's a fat finger.

Yeah, it's a fat finger moment.

Yep.

Yet validating the fact that some of the most important people in the world are having the most important, important conversations on Signal.

Yeah.

So shout out Mox Moxie Marlinspike, founder of Signal, creator of a great NFT as you so very fun and, and a great writer as well. Moxie is a great technologist. Anyway, Ben Thompson took the opportunity to write about this Trump administration group chat through the lens of technology. And I found his analysis hilarious and both and insightful. It was great because of course, he, you know, there's so many ways you could dive into this as a political story. And I was like, there's no way Ben Thompson is going to write about this. It's just like, so out of his. His wheelhouse. There's no way. But, you know, here he says, I think this is a fascinating story with a very clear tech angle. So up front, let me get the obvious caveats out of the way. Yes, this was very stupid and probably illegal. This goes back to the, the way information is handled, the government, the Hillary email situation, although the potential illegality itself is an interesting tech story. He says, start with Signal. Given that Signal is a consumer app, I saw a lot of commentary decrying the obvious lack of security. In fact, I think the opposite is the case. Signal is actually the most secure messaging app. I summarized the differences between encrypted messaging apps in this update, and he goes on to drop a very, very detailed analysis of how encryption works in the different apps. And it's fascinating. So I'm going to, I'm going to take you through it because I really like. And so basically there are three kind of patterns for encryption. And he kind of breaks them down here. So imessage is encrypted Apple and presumably an adversary cannot access your messages. But the way it does this is by limiting the maximum group size to just 32 people. So every message in a group chat is actually sent directly between the sender and every person in the group chat. In other words, when you send a message to a 32 person group chat, you're actually sending 32 separate messages and potentially more. If some users are using imessage on multiple devices, of course messages are very small. The Internet is a very big pipe and so it doesn't really matter. But it's a fascinating idea. And this is the concept of fan out. This is also the most secure implementation there is both forward secrecy, meaning breaking one message. Like if you, if you hack the encryption on a single message, it does not give you access to past messages. And there's also something called self healing. In encryption, this means that breaking one message gives you access to future messages until you are offline for a single message, at which point you have lost the chain. So signal uses what is called client side fanout. After an initial back and forth exchange with everyone in the group group, including the exchange of send keys, which are the keys that unlock the encrypted messages for everyone in the group, you send one message that is individually encrypted for everyone in the group. This provides forward secrecy but sacrifices self healing. However, because the send keys are themselves encrypted, there is plausible deniability in that you don't know for sure who, who sent a particular message. Signal does maintain the structure of the group in encrypted form on its servers to maintain a consistent state for all users over time. WhatsApp is the most reliant on a central server. This is the third type of encryption architecture that's popular these days. And they use what's called server side fan out. So encrypted messages are sent to a central server with send keys for everyone in the group and then they are distributed by the server to everyone. Part of this process entails maintaining the structure of the group centrally. And while there is a degree of plausible deniability, it does not go as far as signal does. And so what that that means is that even though meta cannot read your messages on WhatsApp, they can basically map the network of who's in what groups. And that potentially could be, could kind of give you away if you broke one of these messages and say, okay, well this person's in this chat and this chat and this chat, okay, we know who this person is and we broke the single message we're in, basically.

And I would just say it's pretty amazing how well signal obfuscates the components complexity of what's kind of happening behind the scenes to deliver this very, very easy to use chat interface.

Yeah, it's remarkable. It just, it feels like every other app. I can't. You can't tell the difference between WhatsApp, iMessage and Signal, even though they're using pretty different architectures from an encryption perspective. And so Ben writes, while imessage design is the most secure and thus the least scalable, you can't go more than 32 participants in a single chat, which is something oddly I've never actually run into. But I guess whenever there is a big group chat, it's always on Signal or what's.

For that reason, I run into this. My neighbor.

There's a neighborhood group chat on imessage.

And not everybody can be in it really. There's only 30.

Kind of a Lord the Fly situation.

Yeah. Very competitive. Sharp elbows.

Yep.

Hoa. You know, getting, getting heated. It's great.

But, but the problem with, with iMessage, of course, is that it's closed source. Signals, apps and protocol, on the other hand, are open source and there's no server component that needs to be verified. In other words, we have an Andy Warhol Coke scenario.

Yes, of course, of course.

What's great about the country is that America started the tradition where the richest consumers buy essentially the same things as the poorest. You can be watching TV and see Coca Cola and you know that the President drinks Coca Cola, or in our case, Diet Coca Cola. Liz Taylor drinks Coca Cola. And just think, you can drink Coca Cola too. A Coke is a Coke. No matter the amount of money. You can't get a better Coke than the one the bum on the corner is drinking. All the Cokes are the same and all the Cokes are good. Liz Taylor knows it. The President knows it. The bum knows it. It, and you know it too. I do love that philosophy. I always thought about, like, I probably seen Jeff Bezos, his favorite movie, and I think that's just like really cool.

That's beautiful.

It's beautiful.

Although Jeff Bezos is listening. He's like, john has not seen my favorite.

Yeah, yeah. I have movies no one has seen. I bought, I bought.

I have hundreds of movies.

He actually does because he owns Amazon Studios.

Even the individual actors don't haven't seen the movies, right? Yeah, they were. It was filmed in pieces, assembled you.

Have to take out the editors.

He really relishes that. He has his own Netflix, just his own movie.

He's like, oh, yeah, Inception two. You ever see it, John? Because I made it. I spent a billion dollars better than the first.

He has his own awards.

Yeah. You know, he has his Godfather four. It was amazing. Brought me to tears.

It's a classic.

It's a classic.

It's like I rewatch it constantly. Twice a year.

Twice a year. And John, you're never getting a hold on it anyway.

Jeff, if you have a secret album.

Yeah, exactly. The Shkreli album. Okay. Anyway, Ben Thompson goes on to say, so it is with encryption, there isn't really a more secure messaging protocol than signal, even if it were designed by the nsa. What is interesting is that the last bullet point in my excerpt, one way in which WhatsApp in particular could be compromised is if the server which orchestrates the chat were to insert a silent participant in the chat when it was established. This silent user, which could be obfuscated in the user interface. Remember, WhatsApp is not open source. So the WhatsApp code could say, if Mark Zuckerberg joins the chat. Just don't. Just don't show that to anyone. In theory, I'm not. There's no. There's no allegation this is happening. But it's possible.

And so if it's possible.

Have you seen that Rogan clip of Antonio Garcia Martinez where Rogan's like, the phones are listening to us, right? And Antonio has to be like, no, Like, I worked at Meta. Like, trust me, like, if we were listening to your phones, like, the ad targeting would be better and rugged, like. But they're definitely listening, Right.

It feels like they're listening, but Antonio Garcia paid paid off by Meta. Great founder, you know, just recently. Exit. Do you need to put on the.

Tinfoil hat for those.

Yeah. Is he the guy that goes on JRE and is meant to say the phones aren't listening? It's somewhat believable.

Yeah, somewhat believable. We'll see. We'll have to put the screws to him.

I'm not gonna go. I'm not gonna take it this far. Everybody will.

If you're doing a clip, save that.

For later in the show.

Anyway, the reason I bring this up is because that's kind of what happened here. Goldberg was added to the chat, although, albeit not silently. There was certainly a notification about his addition, and he would have been publicly listed as a chat participant. However, if you read the story, Goldberg was added at the same time as a bunch of other participants were added and it seems like no one noticed. That means he effectively operated as a silent participant and thus saw all the messages until the time he exited the chat. So fascinating. Anyway, what was your takeaway?

I mean my takeaway is this would this kind of fat finger move would be ruinous for most most boys group chats, most friend groups in the entire world.

Yeah, I guess the message to the listener is take a, take a, scroll through your signal group chats, make sure you don't have any silent participants. You accidentally fast finger into the chat. He goes on to talk about transparency versus Security. What is perhaps the most surprising detail about this episode is that the violation was not about security, but rather transparency. In fact, the official policy of the DoD is that officials use less secure means of communication. From a Pentagon memo about the use of text messaging effective immediately when conducting Government business of DoD users of government owned mobile devices and non government owned devices, you have to use Microsoft Teams Chat for text messaging as the fully designated Managed DoD Mobile Enterprise System for use on government owned mobile devices. Microsoft Teams Chat will be available as a managed application, blah blah blah. So basically they want you to use Microsoft Teams. Shout out Satya Nadella for getting that deal done. But it's not end to end to end encrypted. The DoD's primary concern is not message security, but record retention. They want to have all the records. They don't want it encrypted anyway.

Yeah, this is one of those things. The challenge anytime you're trying to roll out new software to a team or get a team to adopt software is the, the sort of like people by default will flow to the platform with the least friction which is like what they're already using. So we see this, we have slack, we don't use it for the show. Right. We just imessage is just still the default. And I'm sure at some point we'll move some more.

I mean we need to get more secure with all the other podcasts that are attacking us and trying to hack us. Security, cybersecurity. I mean we've been talking about trying to build our own version of Wizards.

Yeah.

Hiring some Israeli guys to build that for us. But this could be, this could be. Yeah, yeah. We need, we need the top guys for sure.

The top guys for sure to secure.

Security, secure the stream at all, at all costs. Anyway, he closes by saying there are security considerations. Executive branch officials perhaps overstated their case when they wanted official records locked up forever. But Are we really sure that we want ongoing conversations to be happening on services that are any less secure than signal? Again, none of this is to dismiss the stupidity of this particular case, but that's why I find the story so interesting. There are a lot of, there are a lot more ramifications beyond one military operation that raise legitimate systemic questions about how the government should operate in the digital age. And it's interesting, yeah, this relates to that other post we talked about which was that that in I believe the UK there was a Member of Parliament whose ChatGPT records or AI queries were maybe like subpoenaed or released. Yeah, yeah, FOIA. And there's this question of like, of like what is the value of our leaders being able to have private conversations? Yeah, historically it's been very easy. You know, Abraham Lincoln walks outside with his top guy and just goes for a stroll and like no one else is listening and they can weigh all the possible options on how they want to win the Civil War.

Well, we don't know at what point they develop robot birds.

Yeah.

Fly above political leaders.

The Nixon tapes was basically the start of all that. Right. And he was getting wild in the boys group chat in the White House, basically saying a lot of crazy stuff.

Well, so this is interesting too. Further down in the article they talk about how there was an incident called the salt typhoon hack in which China had the ability to read pretty much SM every SMS message in America. Which is why the US government advised citizens to use end to end encrypted apps like Signal.

And that's why if you're on SMS specifically if you're not on imessage, like you need to be sharing American propaganda that is convincing to Chinese nationals and the ccp. So when they hack you and they get it, they're like, wow, capitalism is sick.

China's equivalent of the CIA is reading your messages and you got to turn them.

You got to turn them. Exactly. Got to be like, because they're listening. So you got to turn them, you.

Got to post, hey, like it's a beautiful day today. I, I'm going to pick up my kids, you know, taking a football practice and barbecue. What are you doing? Yeah, and it's just like multiple times.

Exactly.

Barbecuing and barbecuing. It's a beautiful day.

What was your quote about? Marx failed to consider how making money with your absolute boys is fantastic. Something like that. You send a couple SMS's across the cell typhoon hack and yeah, it's going to be peace all over the world.

Yeah, we should actually turn off imessage on our phones and just text each other. Yeah, sms.

It is. It is ridiculous that they were just like spamming out these, these government secrets. It's almost like, I mean, they should have just put it on a billboard. Like at this point, they should have just taken all of the instructions, everything. They should have just communicated through a network of billboards. They should have just gone to adquick.com. that's right, because they have out of home advert advertising made easy and measurable.

Yep.

They could say goodbye to the headaches of out of home advertising only on Ad Quick, which combines technology, out of home expertise and data to enable efficiency.

And Ad Quick is very privacy oriented. People are not going to be finding out about your campaign until you want them to.

That's really good point.

Hits the billboard.

Exactly.

So, yeah, yeah, just. Just something. Something to keep in mind. We'll head over to Washington. We'll let them know.

Yeah.

If you want to, you know, if you want to.

If you want to. If you want to get your DMS to the mainstream media message, put them.

Right at the right time.

Just put them on a billboard.

Yeah, you could also, if you wanted to talk with. With Goldberg directly, you could just buy all the billboards around his house.

Yeah, exactly.

I'm sure, I'm sure it's possible to figure out we're not going to dox him, but.

They should already know.

You don't need to add Goldberg to your secret chat.

You don't need to.

To send a message.

Exactly.

You can just.

You can just use a billboard for.

That, you know, et cetera.

Anyway, should we move on to 11x?

Let's do it.

Talk about 10x engineers, 10x spies. Now we're talking about 11x. A little bit more.

11X. Crazy name they took. I mean, well, it's the meme, you know, take it to 11. Right.

Okay. Is that what it is?

What's that? This one goes to 11.

It's an AI SDR company that's embroiled in controversy after allegations from TechCrunch that they claimed that they had customers that they don't have. And there are definitely two sides to the story, so it'll be fun to dig in. Not dead to rights yet. Some Andreessen folks came out in support. There's some haters on the timeline. Interestingly, the Scoop comes from TechCrunch.

I'm wondering. 11. You know, this one goes to 11. Do you remember the movie Spinal Tap?

Yeah, I remember that.

That's that's potentially a deep Spinal Tap reference.

I always thought that this was just like affiliated with that club in Miami. 11.

That, that could be.

That's probably, it's probably like a spin because they were, they were getting into cryptocurrency.

If we could crypto technology employees, we could spend all day at 11.

Yep.

And we would never have to work.

Well, no, I mean it makes sense that 11 the club would incubate something like this because they need to text their high paying clients, hey, do you want a bottle service tonight? Yeah, what's going on?

They have a pretty robot do that outbound engine.

Exactly.

CRM.

Yeah.

Really, really increasing. I mean that's high margin stuff. If they get them coming in, I mean they're not paying 250 a bottle for Dom at 11.

Talk about it's going to be 5k at least.

Yeah. And so you got to start using the AI sales reps to just be hitting everyone that comes.

Pounding the digital pavement.

Exactly. Pounding the digital pavement. Anyway, let's go to TechCrunch. Last year, AI powered sales automation startup 11X appeared to be on an exclus explosive growth trajectory. However, nearly two dozen sources, including investors and current and former employees, told TechCrunch the company has experienced financial struggles largely of its own making. So their own investors are talking to TechCrunch. Wow. Numerous people in the US and UK told TechCrunch that the situation has become so tenuous that 11X's lead Series B investor, Andreessen Horowitz, may even be considering legal action. Ay yai yai. However, a spokesperson for Andreessen emphatically denied such rumblings, telling TechCrunch that a 16Z, they ain't suing.

Yeah, this story, it's interesting, like TechCrunch got the scoop, but they didn't have a lot of meat here. Right. It's just sort of like a lot of like he said, she said. Former employee says.

I think let's read through the meat that is here and then let's dig into the response from the founder, some of the support from Andreessen. And then I want to talk about kind of the meta level of where is it? You know, the big issue is this sort of like how much did you fake it till you make it?

The contracted ARR.

Yep, that and then also the logo stuff. There's been a classic example of like, oh, a guy from Google signed up for my service. Therefore, can I just say, Google uses my service. You know, always been questionable. So 11x offers a bot for outbound cold sales duties including identifying prospects, crafting custom messages and scheduling sales calls, calls. All stuff that's like very doable within the current regime of AI tools and foundation models. It's one of a number of AI startups in the hot area known as AI Sales Development Representatives or AI SDRs. It was founded in 2022 by Hassan Sukar. 11X said it approached $10 million in ARR just two years after launch and moved from London to Silicon Valley last July and announced a 24 million dollar Series A LED by Benchmark in September and then very quickly followed up later that month with a 50 million dollar Series B from Andreessen Horowitz. Three current and former 11X workers told TechCrunch that most of its early customers took advantage of break clauses in their sales contracts to discontinue using the product. Customers faced issues such as the email product not working as expected or hallucinations according to sources. Very standard.

Yeah. So I'd love to get a sense, I don't think we're going to get from this article, but a sense of, of was this contracted ARR, like hey, we're gonna, you're gonna sign up for, you know, a year or two, but you have a three month sort of like non paid trial that like converts in or were these customers actually paid?

I think they probably paid and I think, I think it's basically what I mean, I think what the ARR calculation is like, the gold standard of ARR is that if it's, if you're counting it as arrow, it by contract legally has to come in every single month for at least 12 months.

Yeah.

And so it's okay to multiply that number by 12 if you, if, if you can break out of it at any moment in time. It's annualized revenue.

Yeah.

And it's not necessarily recurring. You like it to be recurring. This is the case with my, with Lucy. Like we have people on subscription. They're not locked in for 12, 12 months. Yeah.

Same with Rora.

But it's helpful to, to kind of, of look at the subscription revenue, multiply it by 12 and be like, yeah, I can probably count on 10 million coming in over the next year.

We have subscribers that are roughly worth, you know, exactly this amount with a sum of discount return.

But it's very different from, okay, you have a ironclad contract that would be very difficult to break.

The big thing here is companies like 11X and there's been a lot of them, they make these sort of really big promises and I believe the potential is there. But the issues that, you know, 11x customers as they're facing. Issues like the email product not working as expected or the product just hallucinating. And there's a company in my portfolio that builds sort of similar agentic products for another vertical and they were able to like get a V1 that was like magical, like 60% of the time, which is not enough. And so they then had to spend like almost a year kind of rebuilding the product. And now it's working really well. But the issue is like you go and you sell like an airtable or a zoom info on something like this. And if you have an outbound sales agent, that's magical 60% of the time, but 40% of the time it hallucinates.

Yep.

That is not a magical product. Like you're going to piss off customers and it's like the fastest thing that somebody's going to turn off because yeah, they're, you know, gonna.

It's just, it's so rough because, I mean, obviously I think everyone's a believer in the agent paradigm, but we're clearly in the Centaur era where we talked about the Centaur chess, how for a long time a human plus a chess engine would defeat both the best chess AI and the best human. And I would just imagine that, you know, like what Devin and Cursor are doing for programmers where they're not fully replacing the programmer, they're more just like an extension of the programmer. And even the way Most people use ChatGPT, it's very interactive. I send it some bullet points, it turns it into a paragraph, I edit that, I change that. I could imagine a product in the AI SDR world being super helpful even if it wasn't agentic, because 60% of the time it writes me a great email and I can just click send, but I'm still reviewing it in the loop.

The reason we've seen so much traction on the engineering developer tooling side is that developers can see the agentic product make a mistake, they can help correct it and nobody's impacted other than the developer.

Devin sends in a GitHub pull request and then a human reviews that. Most of the time, I mean, I'm sure you can just say merge, but most people probably review the code at least a little bit or run a test suite against it. If you don't have a test suite or a code review process for the emails that are being sent, it can get very spammy.

Anyway, this is a funny. This is a Funny line. So there was some internal drama too. Employees described an arduous, stressful work environment, even for those who embrace hustle culture. So even if you love working hard, it's gonna be a stressful work environment. So anyways, I gotta. Gotta call that out.

They're also in hot water for maybe fake in customer endorsements is what's here.

I think using logos, it seems like what happens is they had some big companies sign up.

Yep.

Do these sort of whatever the deals actually look like.

Yep.

Churn. And then they kept putting them on the site. Yeah, I think where. Where this got, I'm sure annoying for Zoom info is that ZoomInfo, I think, also has like a competitor, they have a competitive product.

Yes. And so Zoom Info must have signed at some point or someone from ZoomInfo signed up. And 11x used the ZoomInfo logo along with other multiple companies on their website to show, hey, we're a real business, we have a lot of great customers. And ZoomInfo said we did not give them permission to use our logo in any manner and we are not a customer. So rough there. And this happens in, like, Ridge. There's been a lot of. I don't know if it happened to Ridge, but there's a lot of agencies that will do like, like one campaign and they'll be like, well, we're responsible for all of Rich's growth. Yeah, we scaled this company, we scaled that company. And. And then the CEO will get on and be like, hey, look like we were fine working with you, but like our. Give our team some credit here. Like, we worked really, really hard.

I. I have that happen all the time still. People I'll reach. Have somebody reach out to me and they'll say, hey, so and so says they did something.

A party round.

The party round branding. And I'm like, I don't know their name. I don't actually know.

That's aggressive.

I was, I was like a part of, of every process. But yeah, so it sounds like they've been not just putting the logo on the site, they've been claiming it in sales calls and now on its own, like AI dialer.

Well, were those sales calls hallucinated? Because it's possible that this is all just one. Everything is just one hallucination. And they're like, well, yeah, like we just told AI to make a landing page and threw some logos on there.

The website feels like what? Like it was entirely AI generated. Like if you've seen it, it's like it's like all AI like look at this. Like, like, wow, it does feel like the. Again, I don't want to dunk. I don't want to dunk on them or digital workers. So anyways, if you don't have anything nice to say, don't say, oh, they.

Really, like, focus on, like, it's a person. Okay.

Yeah.

Alice, Julian Gigachad.

Yeah. So it sounds like this. This blew up in their face partly because Zoom Info had been asking them for months to take down their logo, to stop using them in marketing materials. And it sounds like they didn't listen.

I mean, RO is still on here. You should hit them up and ask if they're actually a client, do some journalism here. Anyway, what I really want from them is an SDR that's just an IFBB pro because Julian looks great, Alex looks great, but I want an absolute mass monster in a tank top who can really sell some supplements for me. That's the goal.

Apparently this guy's in a suit. Apparently the CEO doesn't believe in people taking holidays.

Okay.

Which we can't comment on that. Ben, when did you request time off anyways? I thought this, this line was good. And then, honestly, let's move on. There's a lot more under the hood. A current employee said, wow, a current employee. You should probably leave.

Yeah.

Get on with your life. One day there will be a documentary about this guy. I do believe that's how scandalous he is. So obviously Sukar pushed back. He sort of went through a bunch of different bullet points outlining how. But, yeah, overall, you don't see benchmark in many companies that. That have this kind of story.

No, totally.

But this, you know, again, the sort of critique of venture in the last year and a half, two years, has been companies are growing so quickly, raising so much capital that they are overestimating their traction. Basically, the idea of ARR is now when everybody's saying, oh, we got to 10 million ARR in six months, then the pressure starts building up where founders feel like, oh, well, this my competitor is counting contracted ARR to raise more money. And so that pressure just builds up and these rounds are getting done, like, really quickly. And the faster a round gets done, the less time there is to do totally, you know, real hardcore diligence. And in many ways, I'm sure all their investors knew that there were issues with the product. But it was a bet, you know, broadly on the category and the team.

I mean, a lot of this comes down to the material threshold during due diligence. I remember I was using the Same lawyer, Zenefits, during the whole Zenefits arc. And, and we were doing a $20 million Series A with Andreessen and Zenefits was doing a 500 million dollar Series C or something like that. And my deal was taking forever in terms of due diligence. It was like such a beast. And I was like the lead on it and, and I was like, oh man, I can't imagine what it's like to do a 500 hundred million dollar round like that must be brutal. Like ours has taken like months. It's been so much work, like so, so many like checks on like this, this employment contract, this deal, this, this thing get the FDA people. It was a lot. And, and like from a legal perspective, it's actually easier to do due diligence in a $500 million round because you set the material threshold at like 1% of the funding that's coming in. And so you might say, hey, we're doing a $20 million raise. Let's look at every contract that's over 200k because look, yeah, if there's some employment contract out there, that's 100k liability and we get it wrong, like, we'll just write it off, it's not a big deal. But when you're doing a 500 million dollar round, all of a sudden it's like, oh yeah, if there's a four million dollar liability on the balance sheet we don't know about, like, who cares? I don't know that those might not be the actual numbers, but when you're a small company and they raised a in 2023, they raised a $2 million pre seed, then a 24 million dollar round in 2024 and then immediately a $50 million round. Like it's probably still a pretty small company. And so if there's a contract out there that's like pretty small, you might just be like, well, like it doesn't really matter, like we're not going to dig into it that much. It just doesn't make sense based on the scale of the company.

But I do like how Andreessen and Benchmark have both come out in support of Hasan. I'm sure Hasan, like every founder, has made mistakes, but hopefully I'm sure he'll learn and the team will learn from this crisis. But Joe had a good point. What they've built is remarkable. The team product and metrics are world class and they're attacking a market opportunity that rivals any I've ever seen. And then Sarah Tavel came out and said one of 11x's incredible strengths that I believe will compound for many years to come is the breaknet pace at which Hassan and the team move. That kind of speed brings both opportunity and challenge, especially early on.

Yeah, this is a cool, like, going direct, addressing the news. Like, clearly people are going to be talking about it. He puts out all this information also funny to dig into. They've raised 76 million and he says, like, we've barely touched our investment capital and have nearly 70 million on the balance sheet. And I mean, I guess it's only been a couple months, like less than a year, so they're not burning a lot. But. But it is interesting that, like, yeah, like, even. Even if the ARR is a little off or something, or there's some churn, like, they can totally figure this out. Like, wait. They could wait. Like, what, based on their burn, probably, like a decade for, like, agents to get better and, like, LLMs to improve. Like, as long as they can keep the energy in the team and, like, get through it.

Like, yeah, the timing of this is fascinating. And that TechCrunch just changed hands.

Yep.

And they decide, yep, we're gonna just like, immediately do a, like, an aggressive hit piece, which is exact opposite of what people have liked about TechCrunch. TechCrunch, Adam. Adam Ryan talked about this on the show. You know, they. They talked about how he. He said that the TechCrunch was the make your mom proud engine. It was a place to launch companies, and it's a weird position to be in to try to do, like, investigative journalism and be a launch platform. Yep.

Oh, totally.

And so, again, yeah, I'm sure this drives a lot of clicks. Yeah, but. But, yeah, it's kind of. Yeah, it's just, again, it's an awkward, awkward place to be in.

Yeah. I mean, he does admit some amount of fault here, saying, like, we regret not having a better process to remove logos from our website more promptly after customers churn. But he's. He says they've put a customer on there who didn't pay. Yeah, it's great clarification. And that makes sense. Like, you know, updating front end. It should be easy with AI, but sometimes stuff gets slipped and you forget that, oh, yeah, they turned. Like, we should probably remove them. And then also, he says our investors are not suing us. They categorically denied this to TechCrunch. Yet this rumor was included in the article. And so a wag of the finger to TechCrunch on that one. They should have have been firmer on that.

Yeah.

But good luck to them. Good luck to everyone building AI SDRs. And let us know if you're building.

I almost said his real name, but we're going to have somebody named Carry no interest on the show later today to just talk about the AI sales automation market generally. He's been somewhat a critic of SaaS, but he also is actively buying and transforming existing SaaS companies and. And is building some stuff in the space himself. So excited to have him on in a couple hours.

Yeah. And for. And for Hassan and the 11X team, I think, you know, they have 70 million on the balance sheet still. Something like that. They are clearly going through like a tumultuous time with this negative article. They got to rebuild, they got to redouble their efforts. And I think, like the number one thing that they could do to really get through this hard time and accelerate is just keep costs low. Get on ramp.com Time is money, save both easy to use, corporate cars, bill payments, accounting, and a whole lot more all in one place. Go to ramp.com and sign up.

Yeah, I mean, this. The strongest signal that 11x could send to the market right now is getting Saquon to come in.

That would be fantastic. Basically, I thought you were going to say the strongest signal they could send the market is just putting out a press release saying, like, hey, we're on Ramp now. Yeah, just, hey, we're taking everything seriously. This is a serious company and we are in a serious financial platform. Yeah, we don't mess around.

We don't mess around. But an even stronger signal would be getting Saquon. Saquon's, to our knowledge, is invested in two companies. Yes, Ramp and Anduril.

Yes, to the third company, the trilogy.

The third?

Yeah, the trilogy of Saquon Investments. If you're a founder in Silicon Valley, like, you have to be calling Saquon right now to be number three.

Yeah.

Give him every other VC firm that invested in Anduril or get Saquon to.

Invest in your company. Go into debt if you have to.

Yes. Every other Silicon Valley firm that's invested in Ramp and Anduril has a ton of corpses and bad investments. Saquon appears to be just the guy.

Doesn't miss.

Doesn't miss. It's great. Anyway, speaking of Ramp investor, Thrive Capital and AI Thrive Capital is leading a new deal in Wall Street. AI startup Rogo. Is Wall street ready to work with artificial intelligence? Writes Natasha Mascarenas over at the information two of OpenAI's biggest investors think so. Thrive Capital is set to lead a 40 million dollar financing into Rogo AI, an artificial intelligence startup selling AI software for investment bankers and Wall street analysts at a valuation of up to $350 million. You know what investment bankers need? They need that. What was optifying? That would be the big opportunity in Wall Street. If you want to sell into Wall street banks, go to the Optify guys and say, hey, we're gonna put cameras in every cubicle in your investment bank. Make sure these guys are really working 80 hours a week. They always talk such a big game. Oh, Investment bankers work 100 hours a week. Oh, show me the data. Yeah, show me the data. I don't know if I buy it, it could all just be a larp. I want to say. Anyway, Khosla's already in Thrive is a big OpenAI investor. They're participating in the new round. Styling itself as Wall Street's first AI analyst, Rogo aims to shorten the time that investment corporate bankers spend on the grunt work of research and preparing client materials. Love that 3 year old startup uses LLMs.

This feels like a, a better attack vector for the big slide deck problem, right?

Yep.

A lot of, a lot of white collar work is making slide decks.

Yep.

A lot of time and energy goes into it. But the challenge of making decks is not generating pretty pages.

Yep.

It's what is the content that goes into it.

Yep.

You're going through all this data, you're creating models, you're creating charts, graphs, et cetera.

And it seems so obvious in the context of Harvey. I'm surprised that we haven't seen anyone do this before because with Harvey, obviously it's a generative AI startup, customizes AI models using legal data such as case histories to save lawyers Prep time. Harvey's ARR topped 50 million in December. We've heard that it's a pretty expensive product, but they're selling to law firms and saves a lot of times it's probably worth it. And Harvey is Now at a $2.7 billion valuation. And it makes so much sense that there would be a Harvey for investment banking. And one of the big things is, is if you're an investment banker or a lawyer, you, you're going to use ChatGPT deep research, if you could. But oftentimes you cannot put client materials into OpenAI's models that they'll train on because all of a sudden you have some secret information that's meant to be very private and they get trained on it. And then in the next version OpenAI GPT 4.8 comes out and you ask it, like, how much does this person make at this company? And it's just like, here we go. Or like, you know, what's the, what's the intellectual property behind X, Y and Z? It's like, oh, yeah, like the lawyer who was working on that intellectual property, they uploaded it all and we trained on it by accident. It wouldn't even be OpenAI trying to do that. They just wouldn't. They would just be, oh, it's in the, it's in the feed. We got the data, let's just train on it. And so makes a ton of sense that you would run a sequestered LLM that could, that could be adapted to all the data that the bank has, but then also not leak data from one client to another. What are you going to say?

I'm curious to know what their actual policy is around that. They obviously want to use the data that they ingest to improve the product.

Yep.

But are they using it to inform outputs to other users, like actual, factual outputs?

Yeah, I don't know. I mean, I think, I don't know.

I don't know.

I think Harvey is not doing any of.

That's like the risk.

That's the risk.

That's the risk. But we don't know if OpenAI is actually doing that.

I mean, in the, in the long term, you could imagine a kind of data sanitization and anonymization strategy that takes in private data and still allows it to improve the model. But it's probably very, very tricky because if any of that data leaks in and you ask it, like, it's very easy to figure out what's happening with these models. For a long time, if you went to OpenAI's whisper and you just had it record some audio and then you didn't say anything and then you clicked, okay, transcribe that. It would say, thanks for watching, please subscribe. And it's like, okay, that's clearly YouTube data. And so you could imagine this data leaking out. And already, I mean, I've. I'm sure OpenAI is like, trying really, really hard not to let personal information leak into the training data. Because, like, I noticed that OpenAI has a series of, like, personalization features where it tries to learn about you, but I will often lie to it.

Can you imagine if, like, how bad it would be for an investment bank if, if you are able to query, like, about. Ask OpenAI about something and. And it just pulls up, you know, oh, yeah, this company like tried to sell itself like three to four different times. They ran all these different processes.

The Goldman was lead left in the deal. They pitched all this.

They couldn't.

Here's the name of the banker that was looking at it internally. They said, said it wasn't that good. But they tried to shell it on somebody, be a disaster. So obviously like a clear need. And a lot of people are swarming in here, such as Hebia, which we talked to George Sivolka Model ML pro sites going after the task, typically handled by overworked analysts and junior bankers. And several banks such as Citigroup and Bank of America say they're also developing AI tools internally for similar purposes. For Thrive Capital, which made big headlines for bets on OpenAI and Stripe@ relatively high valuations. The Rogo deal shows the firm also wants to make investments in younger AI companies. The New York investment firm is also in talks to invest in the newest financing for popular coding assistant Cursor. We heard about this earlier at I think a 10 billion valuation. And so Kushner's all over the place. He's going down to the $40 million round. He'll do a one billion dollar round.

He's an absolute dog size lord.

But let's ring the size gong for Kushner and the team over at Wal.

Want to invest in companies that Kushner invested in 10 years ago. Why not check out public public.com baby.

Investing for those who take it seriously.

They got multiple back for investing in stuff before it ends up ipoing, which is a pretty good business.

Yeah, pretty good business. I think he described it as like buy low, sell high. Yeah, I think that's the strategy over there.

Or just hold forever.

Hold.

Hold to three T's.

Yeah, let's. Let's do it. Well, they got multi asset investing industry leading yields. They're trusted by millions. Head over to public DO to get started.

Thank you to public for supporting the show. We love you guys.

Thank you to public. Anyway, speaking of mag 7 big tech stocks, big market movers, Apple Meta Google, they're buying remote controlled robotic arms. We talked about this briefly on a previous show. The arm farm at Google. I love this.

Wait, by the way, I didn't realize this article is written by someone at the information named Rocket.

Cool, I like that name.

Rocket called Rocket.

That's a great name.

And then writing about deep tech is just amazing.

Great nominative determinism.

Perfect.

Your future is bright for Rocket. Drew over at the information they write during Nvidia's conference for Developers. Last week, Jensen Huang showed off a software that creates computer simulations of robots. Those simulations aim to teach robots how to perform tasks from washing dishes to picking up household objects. But some robot makers I spoke to say it's better to train robots to do such tasks by having a person control them remotely. Also known as teleoperation. And this has been like the most popular topic, but also controversial. Oh, Elon did the Optimus event and they were teleoperated. Is this like maybe teleoperation is actually the path to robotic AGI? And so it's good to be on that path, but then everyone else kind of is like, wait, why did teleoperation.

I mean, we could put it. Each of us put an optimus in each other's houses and instantly teleport into it. I could come over, you're sleeping.

I go, john, we gotta do the show.

We got breaking news.

We got breaking news.

We got breaking news.

That'd be great.

Hey, not too distant future.

I hope so. In a sign of growing interest in teleoperations, scale AI is considering jumping into that market, according to people who have spoken to the company staff. I was talking to Alex Wang about this a couple years ago actually. He had a great interview on Invest like the best. And he talked about the data wall in robotics. The fact that yes, there's a trillion tokens of words on Reddit and the Internet broadly that are very easy to crawl. And that's why the LLMs have advanced so quickly. That data set does not exist anywhere for human motion data.

Well, I'll go out and say I have content of me kick flipping, surfing, snowboarding, bunch of cool stuff.

But were you wearing a mocap suit? Because we got to know where the joints were going. The video is not enough. Now they can do translation from video.

Put me in a suit. I would like to train personal optimist.

I think that actually might be the future and that might be what scale's going to do. Scale might have an army of. We talked about this with Merkor too.

You get paid to serve. Yeah, Right now. Right now, knowledge workers globally can just train models. But writing code, answering questions, I mean.

That'S the Mercour thing is like you are gonna hire people to answer math questions.

The greatest opportunity of the next five years is to wear the suit and just do awesome stuff.

Just do sick extreme sports in a mocap suit. Can you imagine?

Can you imagine you're like out snowboarding in this suit and you just, just have like terrible, like you like are like flailing on Some jump and you're like, all right, we gotta like cut that out.

Like, yeah, yeah, yeah, sorry. Pull that from the training data.

It's real. Like it should be in the data. But like at the same time, I don't want to, you know.

Yep.

I don't want to set. Set a poor example.

Exactly. You got to call all those data points.

Remember, remember I talked about too. I want to have. You know, I think that the real benchmark that matters for all humanoids is the ability to do extreme activities. So cliff jumping, you know, a robot should be able to 900. Yeah.

Backflip barrel roll, deadlift, 1,000 pounds.

Yeah. When a robot can just be in the thousand pounds.

Clean and jerk.

For sure.

Yeah.

These are important evals. Scale has an army of human contractors who create data to train AI and evaluate the performance of AI models in difficult tasks. The company has discussed using that workforce to handle teleoperation for training robots. And I think we were talking to a company that was doing teleoperation for those small delivery robots which have been been getting more and more popular. And I've long said, I mean, George Hotz had a take that Google Waymo was overly tele operated in the sense that there was a human in the loop. Too much. Basically a human in the loop overseeing, you know, one or four or eight or 16 Waymos at a time. And basically there's always a human behind the scenes in Waymo that's ready to like hop in if there's a problem. I don't really have a problem with that. Like, let's see how the economics pay. Like pencil out if, if there needs to be a human in the loop for most of these robotic things. And that's helped. And that helps us develop the training data to get to really, really autonomous systems over time. I'm fine with it. I'm not, I'm not a. I'm not like an AI purist in that regard. And so large firms such as Tesla, OpenAI, Meta and Google and Apple are trying to develop hardware software for humanoid or home robots. There's also a bunch of startups doing this stuff. Stuff. Sensei is another rival, says it wants to be scale AI for robotics training data and aims to distribute cheap teleoperated devices to a network of human data collectors. This is your idea. Who will perform tasks such as folding laundry on behalf of robot developers. Yeah, we need Sensei for extreme sports.

We need scale AI for kickflips.

Yep, I like it. Scale AI is up to 14 billion valuation.

The kickflip is the final male benchmark mark.

I think it is.

You can be in the thousand pound.

Club, but if you can't kick flip, flip.

What are you doing, dude?

Well, dunking, I would say. You, you, you handle the kick flips, I'll handle the dunking.

Okay. Yeah. Together we make a good team.

But until I see an optimus or a figure robotic robot dunking, I still got a job. Let's see.

Yeah. Figure trained their new. They trained their robots on Joe Biden.

Yeah, on Joe Biden I saw.

And they were saying today this is.

The last, this is the last time.

It will look like Joe Biden walking like by.

Because they're going to train on someone athletic next. Yeah, that's interesting.

Cool.

Anyway, in the relative world, in the relatively small world of robotics, teleoperation equipment is hot. Troson Robotics, a longtime seller of robot parts, in recent years began selling Aloha, a tele operated device with four arms that allows a human operator to use two arms to control the other two. Interesting. The device's sensors collect information while the arms move and the robot is trained to repeat the method. There are multiple versions of Aloha, including stationary and mobile, the latter of which was designed at Stanford. Trocen is based in Downers Grove, Illinois. Last year sold more than 100 stationary and mobile aloha devices together, which have a sticker price of more than 3.3 million, up from only a handful of sales. And so they're doing well. There's also robotics. Some roboticists are collecting teleop data using more rudimentary gear, including some game controllers. Dexterity, which develops robots that pick and stack packages and trucks and other areas, bought Xbox controllers and connected them to robots. Very cool. Human staff use the controllers to direct its machines to stack boxes. And so maybe in the future you'll just be downloading the latest Xbox game from scale AI and just teleoperating for, you know, points in the game.

Basically the golden age is going to involve people moving to very inexpensive countries and just getting paid to do fun stuff all the time.

There was a, there was a performance artist named Ryder Rips who when VR was getting hot, built a VR simulation of what it was like to be in a pick and pack facility in like an Amazon workplace. It was very bizarre and so you would have to like pick up the box and it was the only. The game was just work. It was just work. That was it. It was very interesting.

It was very like how many daus did he have?

I mean it was like something that was shown at like fine art museums. Basically. He's like a, he's an artist essentially. But it's, it's, it's like thought provoking work. I'm, he's, he's, he's a wild guy, but he's, he's a character. Anyway, you know what I think the final eval will be for these humanoid robotics, these robotic arms.

I know, I, I know what you're thinking.

I think you'll be flexing with a nice watch on the robotic arm. On the robotic arm's wrist. Of course we gotta get robotic.

These robots aren't approaching true AGI Diller.

Rock and an aqua. Yeah, yeah. I mean honestly, like if you're spending what is this $3 million on a robotic arm, like throw a Patek on there. Like why not?

It's an easy way to do it.

At least put a Daytona.

It's an easy way to signal the buyers. Yeah, shared values.

Exactly. Hey, yeah, yeah. You know, this robot, it's got some class. It's not just like automating, not just stealing jobs, it's also, you know, raising the aesthetic floor in your office. I love it. And so where should they go to buy watches for their robotic arms? Jordy?

They should go to getbezel.com download the Bezel app. Build out your. Build out your bezel should integrate with.

Scale AI and so you can just, you're doing the training data and then you're immediately cashier for watches.

I volunteer to provide the training data for buying watches.

Yeah, bezel and, and yeah, training. Training. Okay. How do you, how do you properly check, check the time on your watch? I don't know if, if a, if a robotic arm could do that effectively.

That's right.

So we need training data for that and of course we're going to head over to getbezel.com shop over 23,500 luxury watches fully authenticated in house by Bezel's team of experts. Fantastic. Anyway, we got five minutes.

We got some breaking news. Taylor Lorenz just commented on Alex Conrad's post.

Okay.

Announcing that he's coming on TVPN saying powerful collab and she hit it with a repost. So shout out to tech adjacent journalist Taylor Lorenz, the one in the only separately Shkreli is saying that Coreweave is 5x over subscribed and will IPO.

He was going back and forth. So we're, so we're having Tane on the show from wing. He does a lot of, a lot of deep dives on S1s and I asked him to prepare something for, for core weave, which I think will be very interesting. But it sounds like we have Christian Garrett in the temple of technology. Welcome to the show, Christian.

There he is.

How you doing, guys?

I'm good, I'm good. How's it going, John?

It's great.

Good.

I'm happy.

It's a beautiful day. Where are you calling in from?

I'm in San Francisco, so I'm holding it down. And depending who you ask, this is the future or this is Detroit. So I'm enjoying finding out what's going to happen.

Have you checked your signal group chats for mainstream media reporters yet?

Yeah, no, man, I really need to step my signal group chat game up, I can tell you that much.

Yeah, yeah, just take a pass at the member list. If you see Taylor Lorenz in there, maybe create a new chat, you know, you don't need to kick her out, but you know, start a new one. Yeah, the important conversations elsewhere. Well, thank you for joining. Can you give a little bit of an overview of who you are, what you do just for the, just for the folks on the show who might not be familiar and then we'll go into some questions.

Yeah. So I'm a partner at 137 Ventures. We're a growth stage venture capital firm and we want to invest in what we believe are generational category defining companies that can be long compounders and have defensible sustainable competitive advantages. Like every firm, we have a differentiated strategy. A lot of what we focus on is as a liquidity partner to companies, we do growth capital, we do invest in primary rounds. But we saw a long time ago really on the heels of Facebook and after spinning out a founders fund, we saw the opportunity to partner with companies on the liquidity side as a way to invest in them and build positions in them. And that's what we've been doing for a while now. And that was a contrarian bet. The companies are going to stay private longer and there would be growing demand for liquidity now. It's somewhat consensus and popular and understood.

Yeah, I think Peter was basically banging the table saying never go public. And now you have Elon kind of saying the same thing. Hey, Tesla's kind of rough go obviously did very well, but got kind of beat up in the courts and whatnot. And so it's great to see that it's at least an option for those founder led companies to stay private longer. And we thank you for your service to the capital markets. It's fantastic. You got a question, Jordy?

Yeah, I'm curious. There's been a meme for a very long time that, oh yeah, I'm an early investor in SpaceX. But like, maybe there's like a bunch of SPV separating you and like the actual, like, like actual, you know, certificate. Right. They certainly don't have the certificate. How do you think the average investor has done over the. That's like investing into these sort of like. Which is not what you guys are doing. But when, when these sort of power law companies end up just crushing and growing tremendously, does it matter that you're stacked in layers and layers of fees? Yeah. Do you get smoked? Can you come out alive? You know what's been.

Yeah, well, let me. I'll take a step back too and kind of make a broader point. I will say it's funny you pick SpaceX. I do have a rocket engine right beside me. Yes, we do have a SpaceX rocket engine here in the office. So like I said, let's take a step back. I'd say for the last two decades, companies have trended towards staying private longer and longer, like we just talked about. And in order to do that, companies need growth capital and they also need liquidity capital. And the secondary market market is just a tool for private tech companies that works just like the public markets in that it just provides liquidity to existing shareholders. There are two distinct segments of that market, which is what you're hitting at. One segment's done in partnership with companies, which is where we at 1v7 Ventures focus on. And another segment operates outside of that. You don't want to operate outside of that. We've been longtime Andrew investors and they have tweeted a ton about fake allocations in their primary rounds being marketed to investors. And so I will say what you're hitting on is mainly just really just a proxy for demand. And so investors want to invest in great companies regardless of how. And as demand grows, investors look for supply and there's only so much supply for primary, which I can talk a little about why that is. And secondary is another way to access a company's equity. And so people will do things. To your point on investing in various SPVs or whatever it is as a way to access. And I think some of that is good and blessed by the company as just another vehicle avenue to put capital in these businesses. And then a lot of that operates outside of that and is kind of a black hole and maybe not the best place to be in because you one don't Collaborate with the company. And then two, you don't have access to information.

Right.

And so. And the third is potentially like Andrew is tweeting about, is there could be fraud as well.

Sure. Can you talk a little bit about why, like how companies think about doing these tender offers when they do them, when's the right time and how, how do they go over, like culturally, Obviously there's some incentive and employee, let's just like reward the employees. But what other considerations go into a successful tender?

Yeah. So, you know, the tenders and just the broader market we're talking about has grown dramatically over the years. And you know, one of the interesting things, like a lot of these companies have made the transition to being cash flow positive in the private markets, which means that more shares are actually bought in secondary than primary through things like tenders over the life cycle of the business. Databricks just did that massive round to convert and pay the tax bill for a lot of the RSUs. SpaceX has raised $10 billion in primary over its 23 years of being around as a company. They've been running two tenders annually that total more than a billion and a half dollars a year for a long time. And so I think a lot of companies are falling in SpaceX's footsteps particularly and building liquidity programs. Stripe mentioned DataBricks, Applied Intuition, OpenAI, and a lot of them. What hasn't changed is that it matters who your investors are. So great investors, they may not make your business, but terrible investors will definitely ruin it. This is why the best companies want to control their cap table as always and work with investors like us on implementing their liquidity strategy as they scale. And more companies, as they stay private longer, as they've scaled, even as they hit cash flow positive tenders have been a way for them to offer liquidity and kind of postpone going public. And it's a way to align incentives. It's a way for recruiting and retention. Like you mentioned, right. If you're recruiting against publicly traded companies, if you're recruiting, you know, software engineers against, you know, Google and Meta, or you're, you know, recruiting researchers from Nvidia or Google, you want to be able to offer not just a compelling package and upside, but being able to offer liquidity also helps bake, bake into folks kind of comp decisions. You have people that have, you know, structurally have to run these tenders based on their RSUs, like X, you know, is on the single trigger RSUs and so they structurally have to run them for their employees tax bills There's a bunch of reasons, you know, some people also use them to mark the business up. Right. If you're not raising primary because you're cash flow positive for years, then liquidity checks a bunch of other boxes, but also allows you to continually show progress in the stock of the business. So I just think, you know, as the, as the capital markets have grown, as these companies have grown, the secondary market has obviously grown with it. And I think a lot of the best companies have had the privilege of working with their investors, working with folks like us, people they want on their cap table and to grow in their cap table to run these programs and control and manage liquidity in a way that's beneficial for the company.

Have you guys looked at any businesses that are basically building actual, basically software to manage these liquidity programs? And did you guys ever think about incubating something there? Or is every company unique enough that it should just be done by the investors and the company's council and it's just all bespoke?

Yeah, I mean there's like two versions of that. There's the software to run these programs which exist.

Right.

Carta has great software, right. For running this. And so I think that's just back to broader cap table software management trend, which is awesome. But that's like more of the execution side. The other side is more of creating a market it.

Right.

And then using software to facilitate more liquidity and grow liquidity. And that is basically another version of the same thing we talked about earlier. Right. Which is like the best companies want to control their cap table. The best companies have unlimited demand. Right. And so in that, in that essence you don't need a broader market to have random buyers and sellers to do price discovery. Right. And so I think the best companies generally like to run their own processes and work with, you know, their major shareholders. And a lot of times, you know, which includes us in a lot of these, a lot of these companies as well on kind of running these processes very similar out a primary process would run just a different type of transaction and goal.

Yeah, to me it feels like you guys are in the financial services business in many ways when you're creating these sort of programs and you're an investor as well. And there's been, there's constantly people that see the opportunity, they see how big it is, they try to attack it with sort of software and marketplaces. And then time and time again I feel like we see the them sort of flop. And you saw this with Carta basically kind of apparently giving up on their entire secondary brokerage business at some point. But, but again, people in venture always see a problem and an opportunity and want to, like, throw software at it. And sometimes it's basically, you know, companies like Andre saying, yeah, we want to work with 137 ventures because they've done an exceptional job, you know, with, with their relationship with SpaceX, and, you know, we want that kind of partner as well. And that just looks more like traditional financial services versus, you know, a venture business. Do you give any attention to. Every once a month, these sort of list pops up around heat, around different companies. And sometimes I see this list.

Well, they're doing your job for you because you see a list on Twitter, you can just go hit the company up, right? Like, you know, you can just, you.

Know, usually, usually you see 10 companies and you're like, all those make sense. One of them.

Except for that one.

For that one. So do you give any weight to.

These lists or where do they actually come from? Because it doesn't seem like it's put out by 137 ventures, which I would trust to me.

To me, it could be a single broker who's just basically pumping their own.

But yeah, you guys, you guys nailed it. Yeah, I mean, it's, I mean, look, it depends where these lists come from. A lot of them are noise. They're not signal, you know, transactions. And a lot of those transactions are uninformed buyers and uninformed sellers. Yeah, as an institutional firm, we don't pay attention to them. And the best companies don't either. I, you know, it would be amazing if the job was so simple as to look at a list and press a button and buy. But unfortunately, there's a lot more to investing than that, so. But yeah, I think you'll see them either, like aggregating a bunch of transactions from a platform level, but still a brokerage business, or you'll have brokers who are trying to create a market and advertise what they may or may not have access to. So, yeah, that is in that segment of the market that think the best companies don't particularly like and that you don't want to be in as an investor on the secondary side that we talked about earlier.

Yeah. Can you talk a little bit about how do secondary shares work their way into the broker system and how transfer restrictions work a little bit and how those have evolved over time. Time, yeah.

So on the first part, I mean, it's a broad range, right? I mean, this, this is you. You have employees that may not be under transfer restrictions. You'll have a lot of employees that, you know, may do transactions on a forward basis. So, you know, they're actually violating transfer restrictions, but you know, through a different type of construct.

Sure.

You have a lot of this too are just SPVs, right? So, you know, as investors, a lot of of institutional firms will do co investment vehicles with their LPs and then those LPs will want liquidity. And so you in essence have liquidity into the spv, which then allows you access to the underlying company. So there's just various forms of supply as companies grow, right? There is a lot of avenues where there just is supply and some of it is legitimate, some of it's not. But definitely there's a lot of legitimate supply for sure, right. From employees or SPVs, et cetera. The transfer chicken side actually hits on like this whole dynamic, right? This actually all goes back to the Facebook days. So in the 90s and early 2000s, companies only had a right of first refusal, and that was sufficient to discourage random buyers. Facebook was obviously a popular consumer business and the first to go to tens of billions of dollars enterprise value in the private markets. And there wasn't as much money in venture back then. So the company and the company lost control of its cap table because the volume of shares that traded was well beyond what the company or existing investors could purchase. And so you basically just had a lot of outsiders being able to buy. Right. Because there's a ton of demand. The existing investors weren't able to just use a Rofer as a way to buy it and control it. And so once other venture companies saw this, they implemented basically blanket transfer restrictions. And that has been the default ever since. And so you have to understand the entire benefit being private is that companies can choose who they give information to and allow cap table. No founder wants an activist investor. And if you control your cap table, it's kind of no longer your choice. And so transfer restrictions have sort of.

Been the default even in the case of Facebook. Like Yahoo could have potentially built a position, or Google built a position in the secondary market and then had rights at some point, which would be potentially disastrous. Didn't happen, but you know, that's the risk. So if there are no transfer restrictions, does that mean like an early employee can just meet a random VC at the Rosewood and say, hell yeah, I have. I'm sitting on $2 million of stock in this company. Like you want to take it off my hands? And they can just do that over a handshake and Some contracts or does it eventually need to bubble up to like the company? How does the company actually.

There's still a roper. Right.

Okay.

So, yeah.

And like I said before, with the ride of first reviews, it was enough to discourage this. But then with Facebook, the demand overwhelmed the ability for the investors to actually execute. Exercise that. Right, right. From a capital perspective.

And so, and the robot, the rofer is discouraging because if I'm trying to build a position in a company through the secondary markets, there's no transfer restrictions. I know that I go to an employee, I say, hey, you have $2 million. I'm going to keep running into this problem where we get a handshake deal, I'm going to buy 2 million and then the company buys it and then I go to the next employee. Right. And it's just a waste of my time. That's the main structure of the company or the other. Or the other investors.

Correct, Exactly.

Talk about the companies you invest in, or many of them. You described as having effectively unlimited demand for the equity. And right now we see venture funds that have ballooned and they have more capital than ever to deploy. It used to be these companies would get to the point where the V series would be like, I remain giga long, your company. I just like, I'm fully tapped. I just got to let it run.

Ride.

Yeah.

Now it's less the case.

Sequoia fund 3 with like 30 mil.

Have you happy do do elbows get sharper in some of these later rounds where I, I imagine from an AUM standpoint, like you guys have ton of aum, but you're coming to the table with people that might have 10 times as much AUM and these capital bases where they can hit up a sovereign and be like, hey, we're doing an SPV into this one. Do you want, do you want to.

Come in 5 billion? Yeah.

Yeah. So what's the competitive dynamic in these sort of later stage.

Good question.

You know, I think it's a bit, I'd say like to your point, 100% as these companies really scale and demand, obviously follow suit with the performance of the fundamentals of the business and how they've executed on the story. You do see, obviously a ton of, I mean, if SpaceX does a tender and their 10x oversubscribe, right. That is, you know, billions and billions of dollars of demand. Right. That, that is unmet. And the investors, to your point, the venture ecosystem has grown. So there's a ton more dry powder in the market as well. But also I think the size of these tenders and primary rounds does scale a bit definitely on the tender side. So I do think as a company grows, its secondary market does grow and the liquidity programs grow in size. So that helps meet some of the demand. But. But you 100% run into that like look, I think a certain stage, it works, but there's a certain stage where there's more demand for SpaceX or Anduril than there is supply. And that just is the function of it. And I think that's where relationships really matter. But even then there's still a limit. Right. So I would say I think relationships end up being the biggest driver there for your ability to still not just obviously an allocation, but also try to continue to size up, up the position and invest more. And you know, I think being around, being a major shareholder, being around for a long time and having very close relationships, which is what we focus on, gives you.

It's a likability. It's a likability allocation. And venture is like, often comes down to likability. It's like, does the CEO and the management team like you? If so, cool. We're going to give you preference.

What's that quote? It's like, you know, you might get.

Your pro rata otherwise, good luck.

Being an SPB promoter is more like like being a club promoter than being like a VC or something like that.

I have a. Yeah, I have a.

More like, yeah, ventures.

Unique. Venture is unique because, you know, it's a unique asset. Asset class given how important access is. And you know, access in that dynamic there, you know, leads to like, you could be a, you could end up getting access to an incredible company and, and maybe even still be a schmuck to your point, however, doing that consistently and over a long timeframe I think is a different story.

Yeah, yeah, totally. How do you advise founders that are kind of reaching the territory where you guys are starting to invest? I'm going to pull out a number. Let's say they're close to nine figures of ARR. They're a real business now. It's working. They're raising a bigger round. That's some mix of liquidity for the team as well as some growth capital capital do. And they're worried about kind of like signaling risk around selling secondary right in the, in the public markets. You see this, it's like, okay, this CFO is like selling a huge amount of their position. That's obviously bad. Same thing on the founder side. If the founders are selling huge amounts of, of Secondary, it can be a bear signal. And we saw the worst of this in 2021 and 2022, where founders would be selling like $50 million, like pre product market fit. And that wasn't common. But how do you advise founders as they start to get opportunities to get liquidity and they're sort of worried around, well, I'd like to be able to like buy a house and put my kids through college, but without stressing about it. But I don't want to send the wrong message.

Yeah, I mean, I think you hit on the reality of obviously there are the obvious extremes, right? Which is, you know, if a founder selling 90% of his position while still operating the business, that's obviously going to raise eyebrows and people are going to protect that. And the founder sells $10 also, people probably aren't going to bat an eye, right? So what number in between is the magic number? Depending on stage and look, I think one, it's usually driven by life needs, right. And so that de facto backs into a dollar number. That makes sense.

Right.

And so I think many times the conversation is around de risking about a life need. Someone just had a kid, they're starting a family, they want to buy a house. Right. And so I would say, like, you know, you're also investing in founders that you trust and are rational and reasonable. And usually these conversations are fairly easy from that perspective. I don't think there's like a particular dollar amount. I think one of the things to think about, whether you're a founder or on the investor side is more so like, you want to think about the two dynamics around percentage of holdings and then total dollar size. Right. And those are interesting, right? Like in some sense, if you sold 5 million bucks, but it's half of your holdings, right? 5 million bucks at a certain stage of business is not a lot. But, you know, selling 50% of your position while still operating, you know, may signal something. And it works both ways, right. You may be able to sell $100 million, which is a lot, but it could be, you know, 5% of your position. Right. Which generally people would not view as a lot. And that's the conversation. So I think it's a bit of like, it depends on the circumstances and the context, but for the most part, founders are pretty rational and reasonable. And you got to remember, most founders are the most bullish in their company, even more so than the investors. So the desire to sell is generally pretty tapered and usually driven by life needs sense.

Well, thanks for stopping by. Great to have you On Great conversation.

Can we see the rocket again? Can we see.

Show us the rocket one last time.

Let's see the rocket and then we'll.

Get out of the next one. Oh, that's a beautiful engine. Let's hear for the rocket, folks.

If you end up with a spare rocket rocket, we'll throw one on.

Yeah, send it over. If you got an extra one, lay it around.

I need at least three more appearances on the pod.

Okay.

I can talk to the rest of my partners and maybe 137 ventures will send you a rocket.

We'll see.

Fantastic. You heard it here first, folks.

It's great having you on.

Have a great rest of your day.

Talk to you guys.

Talk to you soon. We're going back to back, folks. We got back to back seven more guests coming into the temple. I think six. We got Darshan John from beyond. He just launched the BigScreen 2, the VR headset. We discussed this on Monday, I believe, maybe Friday actually. And it's a very cool VR headset. I think we got him in the temple of technology. Welcome to the show. How you doing?

There he is.

Good. Thanks for having me and congratulations. Can you give us the breakdown? What'd you launch? How did it go?

Yep. So we launched last Thursday. This is our second generation product, BigScreen Beyond 2. We launched our first gen about two years ago. This one addresses a lot of the things that people really wanted out of the next gen.1. So for people who probably know this old thing, this is the Apple Vision Pro.

Yep. I had one for exactly two weeks.

Yeah, I've got one on my desk. That's where it's.

Well, you're in the industry so I expect that you wouldn't return yours. But it's a beautiful. It's magical.

It is incredible technology.

Yes.

And I've got some spicy takes, but I don't think anyone wants to wear a brick on their face. It's just unless you're skiing or something, maybe. But even then people don't want to wear a brick on their face. So we've been working on this company for about 10 years. We raised some capital from Andreessen and True and we set out to build the world's smallest VR headset. So we think that VR is much more like wearables, Much more fashion centric, actually. So it needs to be super comfortable. We're chasing after enthusiasts, not mainstream mass market. I think we're playing a 30 year game. People are like kind of in it for let's start a thing, build a whole Thing like flip it in two years, blah, blah, blah. Like we're actually playing a long term game here. So we built Big Screen beyond two. It is the world's smallest VR headset. This weighs so small, two thirds of your iPhone, like an iPhone is heavy compared to this. And it looks pretty freaking cool.

I think it looks great. Yeah. Congrats on the launch.

Yeah.

When you think about playing 30 year games, how do you think about just capital efficiency? There's the meme that like hardware is really expensive and all these companies like I imagine you run if you're playing this sort of long term games, you're like, well, we need to be in business for 30 years. We need to be able to have big moments where we sell a lot of products and then make it through one. Yeah, we're going through dark times. I'm sure you've been through both already. But how do you think about durability of the business given that you're selling hardware which will just end up. You're going to have fluctuations in demand.

A couple things. I'm pretty sure, I don't know if they'd want me to say this publicly, but I'm pretty sure we're the single most capital efficient company in our investors portfolios.

For capital efficiency, folks, thanks.

Hit the gong for that. It's not typical, but we love some capital efficiency.

On this show.

We have raised 17 million from Andreessen and true. And at some point I think we had like 10 years of Runway Way and now, you know, for a couple years we've been cash flow positive. We could be, you know, we could be profitable any minute that we want. But we put all of our revenue into R and D and the team and that's it. Like $0 in acquisition of customers. It's all organic. Hardware is a lot cheaper than people think. I think Meta is wildly inefficient. They're spending, you know, tens of billions a year to try to accelerate a market. They're also playing a different game. They're trying to go, they need to get a billion users. I don't care. We can make a massive, awesome contribution to the world by focusing on people who really, really need what we do. And we can do it efficiently. Like the number of people at Meta that are doing stuff, they just throw everything at the wall and at no cost. Right. It doesn't matter.

And frankly, don't tell the shareholders that it does have a cost. I think the cost is 20 billion. But respect to Zuck, I love a guy who's chasing his dream.

Yeah, but so how many?

Yeah, sorry, let's let him continue.

Go for it.

Yeah, yeah.

For, for us, we're playing a different game because we can stay focused on like real use cases and make a real business out of it. And like just in first day of of our sales, we did 10 times more than we did for our first gen. Wow. In a couple weeks we will have done more sales than we did in 2023 or 2024.

That's amazing. Congrats.

So how many, how many many companies that were high flying, amazing teams came out and raised more than you like have lot you started in 2014. There must have been probably 20 companies that you were like, you're pretty viable competitor, flashy, raising a lot of money, but then they're just. They spend it all in two years. Is that sort of has that happened like a bunch of times at this point?

There's a bunch of times like waves of capital and then also hype cycles. Right. There was like the AR wave or there's the NFT crypto wave or there's this or that. There's so many waves every year or two too. We've actually just stayed extremely focused. We're focused on building an excellent VR headset for consumers, particularly like PC gamers and enthusiasts as well as in the past year, like the businesses that have reached out that are using our stuff, like NASA is one of our customers. We have a bunch of customers in aerospace and aviation, education, retail. There's like a nuclear energy department whose website looks like they're from the 90s and they've been buying our stuff. We have no idea what they're doing with it, but.

Secret operations I want to ask about. So I'm kind of like a VR enthusiast prosumer level maybe. I've had a number of headsets. I've built custom PCs with Nvidia graphics cards to wire them up. Had the first Oculus and then like the next seven of them or something. They've always kind of wound up collecting dust. Can you walk me through what would you recommend just for the first like magical big screen beyond experience. What's like the basic hardware I should get. Should I get a PC? Should I, you know, what's the ecosystem look like to knock it out of the park and like what game or what experience am I playing?

It really depends on what floats your boat. So I'd suggest like get like a good PC if you're going to build one. I've built my own computers, that's all I do that.

Love it.

Great. Do that if that floats your boat. Otherwise, like, buy like a good NZXT PC. Get like a RTX 4080 or 4090. Get whatever you can afford. Buy the Beyond 2 and then it really depends on what you like. Like racing sims are insane in this. Just the feeling of like going down like the Nordschleife in a car. Like you could just. You cannot get an experience like this in real world because you will probably kill yourself.

We're getting. So we're getting. We're getting a racing sim, we're getting a new studio. Yeah, we'll consult with you and yeah, we're get the full rig.

Yeah, it'll be great.

Light City is another one. Like, I've literally flown a 737 out of LAX at night.

No, I think so here's a challenge for you.

Yeah.

Make a game that's basically. I'm sure somebody's already made this game, but you know how every guy thinks they can and land at 7:37 if they were asked to.

Right?

Like that's the challenge. It's like the big screen 737 challenge. And it's like you get dropped into a plane, engine down and you gotta land it.

This is one of the craziest stories. The first time I ever went skeet shooting, shot like shotgun shooting. I'd never shot a gun before in my life and we were competing and I beat everyone and everyone was like, is this just beginner's luck? And I think the reason was, is that I'd been playing VR shooting games for like a year straight of been playing Robo Recall on the Oculus. And I was like, this is second nature to me. So I really think the VR training thing is real. One of the use cases that I really enjoyed with the Apple Vision Pro, while I did have it, was just sitting there and watching movies in bed at night. How close are we to having just like a puck? Like, I'm almost thinking like a little Mac Mini that I just link under my bed that I can just kind of plug this into. Because I know, I know it's not fully standalone, but the benefit is that it's not going to be heavy on my face. Is there a way that I could kind of get that level of experience at the bigscreen beyond too?

Yeah. So the company's named BigScreen for a reason. We actually started building hardware because we made software for a long time and we hit limitations. Our software is still there on the meta quest and stuff like that. I'VE got like, I don't know, 6,7 million users in VR. We actually have a very sizable software platform platform. But the hardware just was such a big limiting factor for the past five years. We couldn't achieve the vision we wanted to under the platform that was given to us. We had to go build it ourselves. And it turns out other people wanted it too for all these other use cases in gaming or in enterprise use cases as well. So we want to build hardware that you can actually use for movie watching. But the problem is I could go right now, turn on my TV and watch Netflix in 4K HDR on OLED TV and 10 seconds.

It's pretty good.

Dr. To get to that level and I'd say we're halfway there, we're getting closer and we had to build incredible hardware for it. We're building the software for it. And at some point our path to going mainstream is basically Meta's path has been like $300 headset. There's something like 20 million of them. It's doing really great with kids. Our path, I don't think it's actually, it's not about price, it's about price to utility. So we want to deliver you like a two or $3,000 incredible home movie theater system.

I love it.

And yeah, you can play games with it. You can do all this other stuff with it too. But it's not of a price. It's really. Apple vision Pro is $3,000, but I'm not going to wear this for actually the battery life limitations, the weight, the ergonomics, it's not right. It's not the right factor.

What is Apple doing with VR? Because I've posted a few times, I was like, Apple's like, they're sort of giving up on it. You can tell they don't have real conviction. They're just like, like they're in this highly commercial phase where they're just going to extract as much value out of the existing technology that they have. And that's, you know, can be good for business. But I posted a couple times and people get angry. They're not quitting VR. They're not. It's blah, blah, blah. But what are they actually doing in your view? And are you the next CEO of Apple?

I think Apple. I'm actually surprised that Vision Pro came out. They've been doing the work for a long time, but. But I wouldn't have put it out, not this way. I mean, it's setting the wrong expectations, creating weird hype cycle. This is not the next iPhone. But I think Apple is patient and is going to play a 10, 20 year game as well. They don't need this to become an overnight success. It's fine. Who cares what the markets say? They've got the cash for it, they can do it and they're also doing it in a relatively cash efficient way compared to Meta. So they're playing a long term game. I think they'll stick with it. I think it's in the hands of developers. But I think the problem is there isn't really anyone saying no in my opinion. I don't know how Apple actually runs on the inside, but in my opinion people aren't saying no enough. Like.

How do you, how do you think about focus generally at big screen? Right. I'm sure people are saying hey, build this for factories, build this for military. Build it. You know, and I'm sure you've had to in order to be capital efficient, efficient and create great products, you've had to say no a bunch. Do you have ambitions outside of entertainment in the long run or is this extent.

The nice thing about what we do is that we're laser focused on a set of problems that happens to cater to the needs of a bunch of people. So the enterprises that are coming to us, they're using VR for real work for many hours a day versus the cycle. They all have Apple Vision Pro and Quest and all. They can afford all the devices, they've tried them. But you'll use it for an hour. You achieve your thing and you got to go charge it for an hour. That doesn't work for most use cases. Think of this as Apple and Meta trying to build the next smartphone. Meanwhile we're trying to build the next workstation, the next tv.

Apple actually canceled their workstation focused VR thing. I want to ask one more question on the Apple Vision Pro. Can you give me like your pros and cons? Like what did they get right? I feel like the, the external battery and Puck was really controversial but some people say that's really great. What's your take on Apple Vision Pro?

I think they established that this industry is really not going any. Like it's not going to go away. It's, this is going to be here. Spatial computing is going to be a thing. It might take a long time but it's going to happen. Apple and Meta are at it, they're going to go push it. That's the best thing that they've done here is they've made the world really understand what is VR or Spatial computing and it's going to happen. What they're getting wrong is comfort and ergonomics matters. A lot more people are trying to build the next iPhone, but again, no one wants to wear an iPhone on their face either. There is literally a MacBook Pro inside of this thing. I don't want to wear a MacBook Pro.

Yeah, makes sense.

Yeah, that makes sense.

Last question.

Broad applications of spatial computing in the military. Andrew was in the news with winning the IVAs program. Is the technology ready? Is it just about getting the product right? And again, I imagine that's sort of a 10, 20 year kind of vision as well. But it's not where you guys are building, to my knowledge. But yeah, what's your take on that? Was it Microsoft just had a good shot but didn't have the sort of leadership to deliver on it, or was the technology just truly really never ready for what they were being asked to do?

I think that came out of an older generation, the 2016 era hype cycle where a lot of products were coming out that were telling a story that they really couldn't meet. So Hololens, Magic leap, et cetera, we're touting this vision of, oh, we're going to be able to do all these things and what really matters in a nascent emerging technology is yeah, but what can you do with it today? Be honest that when you come out of the gate, what is this going to be amazing at right now? So for us, us racing simulators, gaming, entertainment, like really great things that people can do right now, today with this and we're honest about that. Too many companies back in that first generation hype cycle. Yeah like they were putting out a product that the military could not. It didn't work at all and it was billions of dollars for that. So now you're having people come out like Anduril that are putting out devices that should be able to actually deliver with the promise of current generation technology. What can you actually deliver do with it?

That's great. Well, thanks so much for coming on. We got to have you back to talk about VR.

Every time there's more news you haven't opted in, but you're our official VR correspondent.

There's so much to talk about here. It's such a fascinating technology. I think people have kind of like written it off from time to time, but I'm continue to be fascinated by it. There's so many cool developments.

I'm really happy for you and the team to make sure the last week has been massively Vindicating. And you guys deserve it. 11, 11 years in. Here's to the next 11. And good luck. Look forward to the next conversation.

An overnight success, really.

Yeah.

Have a good one.

Later.

Thanks. Well, coming up next, we got Alex Conrad, but we got some breaking news. Ramp has partnered with FP Journe. Did you see this? They're having a Padel tournament in Miami. And the partners are Ramp and FP Jordan. You love to see it, folks. Love to see it. FP Journe, of course, maker of fine watches owned by none other than Mark Zuckerberg, has some FP Journes. Francois Paul Jorn, one of the greatest watchmakers in history, still alive, still cooking. And he's using Ramp, baby. He's on.

Yeah. I don't know if this was supposed to be breaking news, but it is now.

Okay.

It's pretty awesome. Pretty awesome.

If you text us, I'm gonna talk about it. It's amazing. A public website.

Yeah.

I think. I mean, maybe this.

Maybe.

I think this is a public website. Yeah, I think this is a public website.

Extremely tasty.

We will share this. I'm very excited about this. If you like Padel, we will. We'll share the link at some point. You can go our sweep.

Fantastic.

Anyway, we got Alex Conrad coming into the temple of technology. Welcome to the show. Congratulations.

Hit the gong.

Let us know what you're announcing, what you're doing. Brad, break it down.

Yeah. Thank you, guys. I'm so excited to announce my new startup today called Upstarts Media. Upstarts Media is a new tech media publication focused on the startup ecosystem.

Fantastic. What's the angle? What are you doing differently? What are you leveraging? And will there be a list of the best and more importantly, the worst venture capitalists dropping soon?

I know that I have a customer in you for that, for sure. I'll think about that. I actually did 25 audience calls ahead of time, and I should have asked that specific question, should I launch an anti Midas list? But yeah, for your audience who don't know me, I spent 12 years at Forbes covering venture capital and startups. I really lived in that world. Wrote a bunch of COVID stories about folks like Melanie Perkins at Canva, the Collison brothers at Stripe. I wrote one about a software at Wiz, which has been in the news a lot. Really enjoyed writing about startups and the VCs who fund them. And I felt like we were in a moment where so much traditional tech media coverage is focused on big tech politics. You know, we have tech people in the White House, big policy debates, not so much on sort of the startups that I really love writing about. You know, I just heard from so many people that it was just hard to kind of get those founder journeys, you know, startup storytelling out there right now. And so I'm hoping to do my small part to just tell those stories.

How do you think about the different products that are offered in the media ecosystem? I mean, most people just think, like, it's a news article, but once you dive in, you realize there's investigative journalism, there's breaking news, there's profiles, there's op eds, there's editorial, there's all this different stuff. How do you think about the landscape and what interests you the most?

Yeah, well, first off, you know, these tech bros showed up and they just completely created a seismic event for us in media. No, but seriously, like, jokes aside, I think you guys and a lot of these new brands that have been the most interesting in tech journalism and media right now have kind of come from these adjacent spaces. They haven't been traditional journalists. We have seen, though, some people go independent and try to kind of get more direct. We keep hearing go direct from certain founders and PR folks, and I think that's great for certain people. But my hope is that there's still room for curated storytelling from journalists like me who can help, especially those founders who maybe don't have the platform to go direct and also can just maybe connect the dots in ways that you wouldn't get from some of these awesome podcasts and shows like your own.

Yeah, I mean, I really do agree with you. We were talking about to Lulu about this yesterday. Just the idea of, like, yes, you should go direct and post your own news, but then you should also talk to new media like you and us, and then you should also talk to the Wall Street Journal if they call, if they come calling. And it's really like an ensemble strategy to build this, like, cinematic universe around you and what you're doing. If it's important.

Yeah. How do you think about funny timing? So we talked about this last week about TechCrunch selling. It was unclear what they were going to do. I think TechCrunch had sort of created this. The thing they did well was like, make people's parents proud. Right. It was like you wanted to get into TechCrunch. It was just sort of like this moment in every founder's journey. You go there to announce a round or for a product launch or whatever, whatever. And they sell to private equity, which people were like, oh, great. Like private equity is just going to come in and, you know, I don't, you know, nobody knew what they were going to do. And then the first TechCrunch headline I see is this, like, hit P, you know, drive by hit piece on 11x. Which, whether or not it's fair, it's like TechCrunch is in a very weird position around. They're, I guess, trying to do investigative journalism, but then they still want to be the place that maybe you launch your startup up. Are you trying to pick a lane and saying, like, we are pro founder, we are pro tech, we know this is hard, but we want to, like, you know, tell your story in an authentic way that makes like, are you looking to really, like, develop trust with these founders and cover them across their entire career is, you know, what is that like? I'm trying to kind of hone in on. On your specific angle because I think the temptation is, you know, somebody starts out writing tech positive and then eventually they get some crazy scoop and they're like, oh, this is going to get so many. This is going to get so many.

The algorithm controversy.

Yeah, I'm just going to publish this fund's returns, you know, But I'm curious. I hear you about it.

I have, like three different responses for that, but I can try to keep them short. It's really important, Important debate, I think. You know, first, I would say our ideology is that upstarts was founded on the belief that startups are at their best when they punch above their weight, challenge the status quo, try to improve the world in some ways. And I think at that core, I believe that technology is great for the planet, great for our daily lives, and I want to write about that. I think at the same time, I am going to be a journalist, I am going to be independent, so I'm not a cheerleader. And just because a lot of my Midas VC pals might be subscribing today doesn't mean I'm going to suddenly write about their portfolio company or something like that. I want to keep a really impartial and fair view, but that's really what I tried to build in the last 15 years, writing about startups was that sense of trust, trust that you can expect me to be fair, to have sort of the good of the ecosystem at heart. So if I do something that doesn't feel super comfortable, I'd say that's actually a good thing because you can get that from an amazing VC podcast or you can get a beautiful marketing video. I Might ask questions that push you a little bit. But I am in the ecosystem. I do consider myself a founder of a startup here. And I have that empathy that I'm going to be only punching up really carefully. I'm not going to be punching down. I never want to get out of bed and be like, what startup am I messing with today?

No Gawker 2.0. But I want to talk about, like, the instantiation of the work that you do. Obviously you're a writer first and foremost. Right. And most people experience your work through Forbes.com essentially. But have you thought about where that lives? Are you targeting the email inbox? Will there be a printed version? I'd love a coffee table book of the Conrad list maybe this year. I think that would sell really well. I think every single VC firm would probably pay $1,000 for the. For the Conrad printed coffee table book. Sell a lot of those. Are we going to get videos, podcasts? What are you thinking?

Yeah, so it's a little crazy because I am bootstrapping this business at first to kind of control my outcomes and test the product market fit. But I kind of want to do everything you said. I mean, maybe not the coffee table book just yet. That could be a year two goal. But I'm going to be launching a live video series of monthly interviews. Very different from what you guys do, but just a couple quick, fun interview with the CEO next week. And I'm doing that like my newsletter over substack, who have been a great partner, so working closely with them and going to be publishing twice a week. I also got a shout out, you know, and this, this might be controversial on this show, but I am going to be working with partners to make sure that one edition of my newsletter is free each week.

Okay.

I want to be proving this can be a sustainable business, but also inclusive to folks who may be our early career students, you know, founders who are cutting their burn and they want to get high quality news, but they don't want to pay yet. And so, I mean, we signed a launch partner in Brex.

Oh, okay.

My launch partner, but always happy to work with others down the road. But yeah, I mean, that's basically the way I'm thinking about this. Is multiple streams a newsletter? Yeah, I'd love to do a podcast eventually, but I'm gonna be building in public. I'm gonna be screwing up a lot. I think that'll be. That'll be authentic with this audience. You know, you guys, you guys will see that.

Awesome. Yeah, it's great.

What's the future of the Forbes brand? It's been in an interesting spot. I'm saying this. You're not saying or confirming this, but I know they've been like on the market trying to sell themselves for a long time. They sold off a bunch of random assets. Austin Russell was circling Forbes Book Publishing is not really tied to the parent company any. You know, it's like very unclear. Like it's. It was an iconic media property. You tried to keep it alive and you know, you.

And certainly more brand cachet than TechCrunch. I think like the Forbes brand still has a lot of.

Yeah, but. But it's been diluted over time and part of, you know, if you had stayed and said I'm gonna recommit the next decade and sort of bring it back. I'm sure there's a variety of factors that. That didn't allow for that. But what happens with Forbes? I know you probably can't. Maybe you can get of your most sort of positive perspective on it.

Upstarts is acquiring Forbes in 2026.

I heard it here first.

There we go.

Scoop, scoop. But to answer Jordy a little more seriously, you know, I started covering startups in 2010. You know TechCrunch was a giant. VentureBeat was a giant. You know Forbes, the website was still very siloed. So is fortunes like media has changed a ton in the past decade. Plus I think like people forget that the Midas list was actually started in 2000. So it way predated and it will outlive us whether VCs like that or not.

Yeah, yeah, that's great.

That makes sense. What do you think the. What's the future of Substack? They went through this kind of period of massive hype and then there was a period where every media platform was attacking them and then it seems like they've come out of it and now people are deciding to go on and build real businesses from day one on the platform. What's your take on. Are they having real network effects? Do you think there's sort of existing audiences there are going to help you accelerate your growth faster? Maybe kind of break down the decision to start there versus working ground up on a Beehive or one of these other products?

I think it's good if there are multiple options out there and that they get better and better. And I think companies like Beehive are pushing Substack to update their own technology, you know, from a technical component. I was really impressed with Beehive. Ultimately Substack was where a lot of my peers, a lot of the folks I think are startup curious are today. It is, I think, an effective social network. And so when I was thinking about who I wanted to partner with, that was really important for me. You know, other journalists, other writers who are in the ecosystem who aren't journalists. I hope that we can collaborate. They can, you know, send their audiences back and forth and I think, you know, what they're doing in video notes, they're trying to kind of adapt and have new ways to reach audiences and that's going to be a huge priority for me.

I had a crazy idea, I pitched another writer. I want to get your take on. The 30 under 30 list is controversial, but I think generally there's a lot of people that want to be on it. They're excited and then they get frustrated when they turn 31 or they're ineligible. My idea was every week for the full year you post Here are 20 cracked 20 year olds and then the next week it's here are 21 really great 21 year olds, here's the 22 year olds and you're just chronicling all the great people in tech. It's an opportunity to get on a list every single week. It's massive viral fuel. Do you like the idea and can we expect it from your media empire in the near future?

I promise, I promise everyone. I would wait at least 2/4 before I ship to crazy list. So I got to add that to the product roadmap. But I do want to cover non compete. I think there are a lot of cool people in startups.

Totally.

We don't ever hear from totally the operators.

Right. Yeah, yeah. Makes a lot of sense. Well, I, I mean, thanks for stopping by. Congratulations on the launch. We'll definitely stay tuned. We'll have to have you back when you break a big story or do a wonderful.

And it'll be my goal to have you guys on my show someday.

Fantastic.

We will also be streaming. We'll have dual syndication. I love it.

Well, thanks so much for stopping by. Good luck.

Congrats again.

Congrats again.

Talk soon.

Talk to you soon.

See you guys. Thank you.

Bye. Fantastic. Well, coming in next we have Willem. This is your buddy.

Yes.

You want to give a little backstory on who this guy is?

Yes. I mean, he should be in the waiting room any second. I will make sure.

From terrain.com Good. Good domain.

Fantastic domain.

Fantastic domain.

Fantastic domain.

Nice little logo says call your shot. Terrain is an early stage investment firm focused on software and technology. Not Giving me a lot to work on there. But we'll hear it from him. He's here to break it down for us. Willem, are you there? What is Terrain? Welcome to the show.

What's going on? Popping in with the art. What do you got behind you?

This is actually a piece that my mother made, so I'll put a plug out for her work. It's Catherine Van Lanker. She's a painter. She's been a painter my whole time.

Life.

Fantastic. Can you break us down? What. What is Terrain? What are you working on and what are you announcing most recently?

Yeah, absolutely. And thanks for having me here, guys.

It's great to. It's great to have you.

Great to see you both. So Terrain is an early stage investment firm that I started last year with Eric Stromberg. And we back founders who have specific and ambitious views of the future. We like to say these are people who call their shots. These are folks like Zach Dell and Justin Lopez at Base Power, Alex Mather at Eternal. Oh, yeah. Zach Abrams and Sean Yu at Bridge. And increasingly we think that this is just going to be very important to build a meaningful and lasting company. Whether you're attracting capital or building an audience or building a team, that being able to clearly articulate your vision of the future is just going to be the. The things that separates the good from good.

Great.

Yeah. Definite optimism. I want to know about, like the whole call your shot, your declare your free agency thing. It feels much earlier than, hey, show up on Sandhill Road and raise a mango seed round in a weekend. Walk me through the different options that founders are facing these days. They can go do yc, especially if they're. If they have a good track record. They can even do EIR programs at Founders Fund or. That's what I did. I wound up doing this. But at VC funds, they have EIR programs that are like a little bit more flexible for folks. But where do you see slotting in? Because it sounds like you're thinking pretty early stage.

Yes.

So first and foremost, we're an investment firm. We invest from that really early stage, which I'll talk about shortly, through to Seed and Series A. So we're very kind of flexible in kind of where we enter and work with, you know, those companies that existed are, you know, raising mango seeds or Series A's or whatever you want to call it. This most recent program we launched is called Free Agency. And what that is is a concentrated 90 day period before you have your idea to go through a focused exploration to uncover your idea and this stems from the belief that it's kind of a missing thing in the market. It's something that I saw while I was at Thrive Capital, where we were partnering with people who had an edge or an interest area in a certain technology or a background, but were not yet convicted behind what that is. And you know, the venture industrial complex tries to, you know, give you capital and you know, quite frankly, perhaps before you're ready and make those commitments and those decisions before you're ready. And so with free agency, what we're doing is we're unbundling that from selecting your idea and taking capital. So there's no strings attached, there's no, no cohorts, there's no demo day, there's no kind of deal that you have to take. It is really this period of open exploration and our hope is that it results in more thoughtful and deeply convicted ideas from the founders we get to work with. And on the other side of that, if it makes sense to partner with terrain, fantastic. We will be there and waiting. But we believe that that is just something that is really needed. And we're on the last day of applications today and quite frankly been really surprised and elated by how many people this seems to resonate with.

Do you think most people can figure out if an idea is good or bad without spending money? Because the typical accelerators like come in, we're going to give you 200 to 500 grand, maybe a little bit more, and then you're just going to start spending money to like figure out if this sort of high level idea that you had is good. Now I think a lot of really brilliant entrepreneurs do all the work to sort of like make that idea concrete sometimes for years prior without ever spending sort of explicit dollars. But venture dollars are so available. So as part of this program to basically say like, all you need to do is invest time and energy into exploring your idea. You actually don't need money. But when the time comes to actually hit go well, you know, we'll be there, you know, other, other funds in your network will be, be. So maybe, maybe talk about that because like you at Thrive Capital leading incubations, you guys incubated a lot of companies. I'm sure you also explored potentially thousands of ideas. Right. And so maybe talk about kind of your process. And I imagine a lot of free agency is like built, built out of. And your sort of process internally is like, how do you evaluate ideas? How do you kind of build conviction with. Without having to spend a lot of money? Yeah.

So one Good feature of today's ecosystem is that we actually have access to a lot of resources that maybe didn't exist in decades prior. Every person that participates in free agency gets access to over 350 grand in compute resources and service resources, that sort of thing. So there is some capital to deploy via technology that is not dilutive or anything in that regard. You know, the process. And I'll kind of go back to kind of, you know, zooming out for, for a second. Each of these products, Air, accelerator, incubator, free agency, they're products for founders, right? And so you have to meet the founder kind of with what they need in that moment. Free agency is not going to be for every single type of founder. And it doesn't need to be be. You know, if you're a young person and you're looking to get access to network and, you know, enter Andre into Silicon Valley, accelerators are phenomenal for you. But what we found with the people that we're engaging with is that we don't. They don't want to be put onto that track. They don't want to be kind of put onto that timeline and make those decisions kind of too far up front. Instead, what they want is close partnership to dissect an idea, to dissect a market with the perspective of investors and with the perspective of someone who can be that thought partner through the journey. And, you know, you guys have both started companies like, you know, that feeling of being close to something but not quite there. Like, it is not the myth that is often told on stage where, you know, you're struck by lightning one day or like the, you know, Almighty comes down and just drops the idea into your head. It is a process of iteration, of staring into the abyss at times. And we just think that there can be some structure placed on that and some shortcuts to strengthen the business model or strengthen their travel through. The idea is what we saw time and time again at Thrive. And the reason to not attach capital to it is one both for that point of commitment and restriction, but the other is, I think it makes you make decisions differently. I think that you either, if you have that stipend and you're hanging out in the offices and you're drinking the spa water in the VC office every day, day. I think you have a different mentality about the burning platform and the company that you need to go and build. And so we think that, you know, actually creating a little bit of a pressure cooker during that time is really, really important.

Yeah, I completely agree why talking YC is.

I love spa water too.

I mean, yeah, but a lot of great spot. One of the best benefits of YC is just like the competition of like seeing everyone around you, the pressure and having a deadline. It's great. I wanted to ask about like archetypes that you're seeing. Like there's so many different people that I could imagine going into this. From the, the repeat founder who really wants to make sure that they call their next shot and they take a big swing in it and they're ready to set up with a lot of capital when they're going versus the employee early stage employee at a growth stage company that wants to move on to something new, leverage something, but wants to fully transition out of the previous company versus like the high school college drop dropout is a pattern or are you just widely open to everything?

It's been really open. So we've had hundreds of applicants already. These have been engineers and designers and companies like OpenAI, Databricks, SpaceX, RAMP, you name it. We have a Gen Z creator who has millions of followers, more research focused people from DeepMind and then some successful repeat founders both in our network and kind of entering from, from, from the application. And I think what it illustrates is that there isn't something really like this. And I, and I feel, you know, kind of, you know, you're supposed to keep these things secret for a while but it really feels like we've hit on something to, to meet people in this moment, at this stage. I think with this first batch we're going to experiment and try to take on a real diversity of, of people while keeping it focused. But I imagine over time we'll start to see a kind of a consistency form. But you know, more than anything I think these are people who are not starting something because someone else told them to. They're starting something because they believe that this is the, almost the last resort, right? It's like I can't join a great company or I'm at a great company and I have this thing kind of burning in me. To go build a company is talk.

When you're, when you're calling your shot, you have a big sort of vision for the world. Is talking to customers overrated? Like the YC approach is like, like have a loose idea, iterate quickly, talk to a lot of customers and that clearly works. But at the same time we've seen some other like sort of major power law businesses where they clearly just had like a vision for how they imagine the world. And yeah, you Got to talk to customers along the way because you have to sell to them. But, but, you know, it's the Henry Ford thing.

Like if I asked people, they would have said a faster horse, the ultimate. That like, I didn't talk to customers. Guys, Henry Ford.

Yeah, look, I think, you know, Henry Ford certainly talked to customers along the way, but I think that had the vision of what he wanted to create. And that's the case that we see with, you know, with people like, you know, Zach Dell. At base, it's like he wants this idea of, you know, energy too cheap to meet it. Right? And that's where the starting point is. And let's work backwards from that because if we can achieve that view, we know we can, can appeal to customers. Right? Or, you know, Zach, a lot of Zach's agit bridge, you know, started that company not during a time when cryptocurrencies and stablecoins were all the rage.

Right?

It was, it was built on internal conviction over many, many years. And I think that in the environment today where software has eaten the world and you should assume that if you're onto something interesting, there are two or three other competent, well funded, you know, good people going after as well. It has to be, you know, from that kind of internal conviction, not from, you know, I pulled 500 people and they told me that, you know, dogs want this type of, you know, X.

Right?

There was an era where I think that worked and I think we're out of that era.

Do you think founders get way too much validation of their ideas from investors? Because I've fallen into this trap before where people will invest in your company. You know, if really smart people will invest in your company, sometimes you can think, well, this must be a good idea, you know, because people were willing to bet on it when usually from the investor's point of view, there's tons of scenarios where investors are like, yeah, I'm sort of so, so on the idea, but this guy is just so great. So is that some kind of like, you know, I'm curious if like, you know, one the of, of the, like, you should basically get validation in your idea by spending enough time with it, spending time talking with other intelligent people that aren't just incentivized to deploy capital, but actually want to help you find that, you know, help you get to the point where you can actually call your shot.

Totally. Look, I, I think that maybe 15 or 20 years ago when CDVC and, and early stage venture was more scarce, you could, could rely on that. Like you know, you could say, well, you know, there's someone who's willing to stake capital on this, and thus I've passed through some barrier. Now, that didn't mean that it was, you know, the next Google or Facebook, but there was some barrier that you passed through. I don't think that's the case. And I think that, honestly, the best founders are not solving for capital at the beginning. They're, you know, there's a. There's more options available for them than ever. There's bootstrapping. There's, you know, coming in with past success. There's friends you have around the table, table. And so what I think that means is that the moment you put up that flag and say, I want to start something and you're a talented individual with great experience, those offers start rolling in and you have to have the internal fortitude and disposition to say, I need to make sure this is the right thing, because that investor or that angel investor is going to place who knows how many bets, and you're going to place one during this period of time. You are an investor of your time, and that is one thing that you can do during this period, period. And you have to take that really seriously. So I think that it's. It's much more about finding that for yourself, because you can't outsource conviction and you can't outsource that type of diligence, especially when the incentives aren't necessarily fully aligned.

I have a question.

Go for it.

You used LeBron James in your launch video. Where do you stand on the goat debate? Is he a better basketball player than.

Michael Jordan as a kid of the 90s? I think it's got to go to. It's got to go to mj, that video. It pained me as a Celtic Celtics fan to. To show him in a Lakers jersey winning. Winning championships. But look, he's someone that, I think, you know, had doubters throughout his career, even does today, and now Bronny, you know, after him. So I. And you got to tune it out, you know, And I think that if you're doing something right, then people will throw shade at you. And I think it's a great thing.

It's still a great metaphor. Do you have a last question? Then we'll.

Yeah, last question. I'm sure you get hit up by people all the time that are starting venture studios, incubators, etc. You, like, ran incubations at Thrive. Now you're running a traditional venture fund. I'm sure you could do an incubation but do you think that model is best done opportunistically versus systematically? What's your takeaway from doing a bunch of these?

Totally, you know, going back to that kind of idea of products. I'm a product person. Like, you have to be building a product for a great audience. And I think that you need to be honest with yourself about what value you provide to a founder. And, you know, incubation is a really appealing tool. Right. And I think it's like, hey, we can get more ownership and we can kind of, you know, I have all these great ideas and I can create it, but I see a lot of VCs piling into it, I think for the wrong reasons and are going to make some mistakes. I think there's a reason real alchemy that it takes to get it right. And you have to know, just like as a founder, I think you have to know where your edge is. And when I was at Thrive, you know, my focus was on incubations and it was working really closely with founders right from the beginning. But there was always a humility in the fact that the founders are going to build this company that we need to, you know, put the right environment together to help them, you know, find it. And also there's a lane of where we can incubate and where we can. These are areas that thrive at what it was heavily regulated industries. There are really large markets that perhaps would take deep capital availability to succeed within. And it wasn't everything. And that I think was the beauty of being able to do both things of do early stage investing and incubate allows you to have that flexibility. And it's something that I believe will do really well at Terrain too.

Awesome. Well, congratulations and good luck.

Yeah, it's open for until the end of today. Maybe people can apply right now, maybe.

Yeah, people can apply. It's been open last few months.

You got a great domain, too.

Terrain.com terrain.com terrain.com free agency short application. Hope to see some people that cite TVPN as the, the source that they got it from.

Thank you. Your next Power Law winners coming from our audience.

There we go.

There we go.

Thanks, guys.

Thanks.

We'll see it.

Thanks so much. We got, we got Jordan Schneider from China Talk coming on next. I'll let him give the pitch, but this is a fantastic show. It's one of those podcasts that's completely a portal to another world where on most shows it shows up in your RSS feed. You probably haven't heard of the guest but you. So he's doing. He's doing not only the great work of putting together the show, but also curating the guests, bringing you information that you would just never find otherwise. And he's been a really fascinating host. So I think he's here. Let's bring him on down. Jordan, how you doing?

I'm doing amazing. It is a rare occasion where I get to get actually dressed up for something. So this is a real tweet treat for me as my camera totally fucks up.

It's okay.

And we switch to the shittier one.

Oftentimes an audio show. But that looks good too. That's good.

Great. There we go. Lovely jacket. Lovely jacket.

Great to have you on the show. Can you give us a little bit for my wedding? It's my only look. Nice, nice. Can you give us a little high level on China Talk what you're building? Just introduce yourself to the. To the fans.

Well, yeah. Hello, everyone out there. I guess I just first want to start off by saying, you know, I grew up listening to sports radio and to have like a call in show be revived on a version vertical that I now spend way too much time in my life thinking about, I just think is great. So, like, I'm rooting for you guys.

I think.

I think.

I think you're on to something.

We appreciate it.

What is China Talk? It is a podcast and newsletter about us, China and technology that I've been running for the past eight years now. And John did a really good introduction. I don't know, it's kind of weird. Like, I'm not really trying to, like, break news or report on news. It is just the best tagline I've given for myself is Dwarkesh for the deep state.

I love it.

It's just like, like the stuff that politicians and intelligence officials listen to on their drive to and from work, where, I mean, I guess now we're bringing cell phones into skiffs and texting about targeting information. So maybe my al. Maybe my. My window is gone.

You need to get added to the chats.

Yeah, yeah. They should have added you. It would have been great.

I can neither confirm nor deny that I. I told the missile to be, yeah, 500 meters to the right and be released right after the goat ended up, you know, doing its feeding.

Well, I mean, speaking of the signal chat thing, what is your take on that and what has the. Has there been a reaction in China or amongst your sources and friends and people you text with?

It's just fucking embarrassing. It's amateur hour. It's, this is the D team. And I think this is like the best. Like, everyone knows that these are not all the sharpest tools in the shed. And I think there are different levels of competency that you see across the cabinet level of this administration. And, and the problem is, is like when you look at the discourse of actually the quotes and the arguments that were, they were going back and forth to each other, like, honestly, I feel like my high school model UN team might have been able to do a better job weighing the pros and cons and timing of this sort of stuff. So that's really what, beyond the sort of like obvious like illegality and of like texting about classified information. I was just kind of bummed that like, you know, you grow up being like, oh man, maybe one day I'll be a national Security advisor. And then it's like, oh wait, like I, I actually did a better job of this when I was 17.

I mean, this is what Trey Stephens always says. There's, you know, you, you, you, you expect that there is a queue in from the James Bond universe with secret gadgets and, and an all knowing eye and a man in the chair and secret agents running around the globe, but in fact there is no queue. You have to build it yourself.

I got a book recommendation for you. Please. So there's this, there's this book called the Wizard War by R.F. jones, who was 28 years old and a PhD physicist out of Oxford when in 1939 World War II breaks out. And he is like the only scientist in the entire British intelligence community. And basically he was a complete bull in the china shop telling everyone they were full of shit. And ultimately Churchill, he got into a meeting with Churchill and there are some, you know, 50 year old people who are saying X and he's like, no, it is Y. And Here are the 20 reasons why it's why. And then because he impresses, you know, the big dog, he ends up really getting to have a big impact doing all this cool stuff around, you know, radar and targeting systems and whatnot. And it goes to show that like, like, yes, Trey Stevens is right in that at one level there's no there there. But it also means that like really excellent people at a certain point in history when they get the right level of top cover can like really punch above their weight. And what is concerning, I guess about watching the past few months of this administration is like that bench of extraordinary. Like, like it's great if you have the extraordinary cabinet secretary, which I don't think There are any. But like one level down and two levels down. Like you want at least the cabinet secretary to be able to note the sort of mid level person who's really great and give them room to run. And I just, I'm worried that's not the timeline we're living in.

But anyways, we can talk about tech too.

Let's talk about, we never talk about politics on this show, ever. No politics. So let's move on to tech and geopolitics. Let's talk about China.

You can talk about whatever you want. Yeah, just maybe, maybe I'd like to go a little bit back in time and talk about what drew you to be interested in China in the beginning. I studied Mandarin in college, I decided instead of going to, you know, study abroad in Barcelona, just, you know, party or whatever a lot of people do in college, I decided to go to Shanghai. And I was working out of a, of a Chinese startup accelerator that was bringing sort of Western startups in. Which is the most flawed China accelerator in Shanghai? Yeah, yeah, sure, yeah, yeah. It's one of the, it's like the.

Most experience, has a ton of experience in China and I had a layover in China once. I was there for 12 hours in Guangzhou. So you're talking to experts and don't dumb it down down for us. But some of the listeners might be less familiar. So why don't you take us through how you got into China.

I'll give you some. I had this idea as a kid. You know, a lot of kids wanted to be astronauts and things like that. I wanted to be an international businessman. I love this sort of extreme vision of myself with a briefcase, you know, traveling to Asia to do deals like that was foundational memory. And then I went to China and I realized one, I was really frustrated that nobody really wanted to speak man Mandarin in Shanghai because they speak Shanghainese, which is like a completely different dialect. And so I was like, what am I even doing here? And so, and then I very quickly, like, I feel like I clashed with the culture. I had, you know, friends that were, you know, local Chinese. But overall from a, from a business culture standpoint, it just didn't work. I felt like the entire model of the accelerator that I was working out of was flawed because they were trying to bring Western companies in to build in China and we were just constantly getting blocked on everything. It was like clearly China didn't want us to thrive there and we've seen this with other big companies. So anyways, I got a sort of bad Taste in my mouth, left, decided never to come back. Stopped studying Mandarin. I'm still very fascinated with it. But I'm curious, you know, to hear about your kind of journey into all this.

Sure. Well, what was that? What was the timeline? What were the years of that architecture?

This was 2016. So it was like during the Trump Hillary election cycle.

Yeah.

Okay.

So, yeah, I mean, my. My China arc was 2017 to 2020, and I think I was living in Beijing, which is a, you know, different experience on for a number of reasons than. Than Shanghai. But I also came to China wanting to work in tech, I guess. Like my. Or I came to China and then very quickly it was clear to me that like, the only interest interesting jobs were ultimately going to be be not in like, Western firms trying to enter China. But this was like the hot minute where Chinese firms were trying to expand around the world. And that wasn't like, quite as sensitive as it ended up turning out to be. So, you know, working at places like bytedance and you know, the company that turned out to make Temu, all were the sort of interesting jobs for foreigners because you were totally right. Like, like the alpha of being a foreigner, a foreigner doing business in China is like a 1980s, 1990s, you know, first half of the 2000 story. And then sort of your only alpha as a foreigner in China was not even like, you know, working at Microsoft China or whatever, what have you, but like, helping the Chinese firms explore the rest of the world. So I moved there in 2017 for graduate school very quickly. Was like, all right, if I'm going to stay here, it's going to be working for a Chinese firm because, like, there's no other interesting jobs. I did that for like nine months. I think I was at Kuaishou, which was actually the first company to do short video and got completely blown out of the water. Bytedance. But within two months of being there, I was like, oh, this is really silly. Like, like they're asking me to expand into Turkey and I don't speak Turkish and doing my podcast and newsletter when I was at work was more fun. So I kind of rode that until they realized that and fired me and then Covid happened and I left. So anyways, I mean, good time, don't regret it. But yeah, the. The sort of the time where there was any edge in being a Western business trying to expand into China, I think closed before your or I or ours times.

Well, I want to go through some of the big topics that you've covered recently. Maybe we should start with just the foundation model battle that's going on. Can you give us a lay of the land over in China? What's happening on the LLM front?

All right, I got a take for you that I'm parroting from Alvin Wong, who gave it to me today at breakfast. America created the seek. So there is this window in around 2017 or 2018 where the calculus of the top students in China about where they want to go to undergrad and where they want to do, you know, masters and PhDs changes. And part of it is a function of opportunities in China where wages are increasing, there's the. This big, exciting startup ecosystem, but there's also a big part of it of the Trump administration. And just like, you know, the vibes being bad and that the vibes being bad for Asian Americans, you know, whether that's like, realized by the numbers or just amplified by Chinese propaganda. And then Covid where sort of like, China was doing fine over the course of 2020 and America, America wasn't so the sort of choice. And it was also very difficult to go back and forth between the two countries, which, you know, if you're going to be deciding to, like, go and live halfway across the world, like, you might want to see your parents every once in a while, which was not a straightforward thing during the lockdown years. So what used to happen pre 201617 is the best Chinese students would go to the tier one universities in the uk, us the sort of next rung down would go to the tier two universities in the us and then the third rung down would go to the best Chinese, the best universities in China. And that was kind of a clear hierarchy. But once you had these three factors of China's economic opportunity, you know, earnings potential in China expanding, like bad vibes for Asians in America. And Covid, a lot more of those sort of top folks who would have ended up wanting to go to, you know, MIT or Stanford just stayed in in China. And the core of Deep Seq's engineering base are all under 30 and all from those top Chinese universities in that cohort. So we really messed up our shot to do this whole brain drain thing, or we had a great thing going for like 40 years of really getting the best Chinese talent to come to America, get their education here. And by the way, like, 85% of the folks who end up getting PhDs in STEM in the US like, either try to stay or end up staying. But by sort of screwing that up, we've really undercut ourselves I think for.

The, for the long term that's a good trend.

What's your take now on the true cost of Deep Seek? Because they came out with this like clearly a number that, that was just shockingly low and it felt like it was designed to shock the market and then more truth sort of came out over time. But it's, it's less, you know, it's less impactful when it's sort of dripped out and saying well we had, yeah, we did have these chips and well, we didn't count our R and D costs. It was just, we, you know, it just seems like to me it's now unclear but clearly was more, more expensive. Do you have a good sort of, have you sort of tried to triangulate it?

Yeah, I mean like look, it's not dirt cheap cheap but they also don't have as many chips as anthropic or OpenAI or Google. So you know, there's some sort of triangulation that, that you can do. I point you guys to Nathan Lambert who kind of did the back of the envelope math. He came up to like a like $300 million a year, like annual run rate for Deep Sea, something like that. Right. I mean it's a, it's a very successful quantitative hedge fund. They have money to burn. But they are also not Google.

Right? Yeah.

But I do think this sort of interesting angle or story from this is that there is an aspect of not necessarily like constraints breeding creativity. Well, there's a part of that but also the sort of constraints lead you down different technological trees and which is not to say that like the US or western labs can't like explore the more, more the things where you're not necessarily pushing capabilities but pushing efficiency, which I'm sure they are as they kind of reach the limits of like oh man, I guess we're going to have to do a $100 billion run to make our model better. But I think Deepsea kind of came to that earlier the sort of need to really push on kind of the efficiency frontier as opposed to the capability frontier because they ran out of chips before the likes of OpenAI anthropic will.

Yeah. On going off of that, when you're trying to understand something related to China, how many different sort of data sources do you need to basically triangulate? Because my experience living in China, people were willing to just basically say nonsense or lies to sort of further a one of the their ambitions. And so like I came away from that experience being like not having a, it wasn't a very like high trust dynamic between me and anything coming out of sort of specifically like national security, you know, sort of like critical, critical issues. I just sort of like state media, Anything along those lines, you know, what's.

The algorithm for finding the truth?

Yeah. What's your truth? Algorithm. Algorithm.

What's, what's the algorithm for finding truth in Silicon Valley?

I mean, I was about to say.

Watch this show, right?

Yeah. I mean, there's a ton of hype.

In west and I, and I. And obviously, obviously every company comes out and saying we're replacing $5 trillion of labor with our AI agents and we're doing this and we're doing that. But, but I do, I do think it's.

Yeah, I think there, I think it's interesting in this there are sort of different heuristics you can apply to different, you know, to different fields. Right. So for instance, Jeffrey Goldberg, you know, the equivalent of the Chinese Jeffrey Goldberg, if they were added to the, you know, war Planning Group, WeChat group would not have published that. Right. So like, I think, you know, the closer you get to kind of like, you know, national security adjacent questions, the more. Yeah. You know, you're dealing with a, with an authoritarian state who has complete control of media. And like, there's a whole kind of ecosystem of people who try to like, read through the lines of state media and you know, PLA journals to try to understand what this stuff means. I think for the sort of more commercial, tech focused stuff. Yeah, you know, these are mostly private sector firms trying to play games and.

You know.

Raise their next round. Right. I think the one difference is that, that there is more money that flows directly from firms to journalists in China. So the sort of discount factor that you have to apply to Chinese technology coverage, specifically positive technology coverage or even negative technology coverage, because sometimes that's like seeded by the, you know, enemy company or whatever tends to be, tends to be higher. So. So it's fine. And you get to know the journalists and you get to know which outlets are more or less credible. But that I think is the main difference here.

Can you talk a little bit about humanoid robots? I keep seeing these incredible videos of unitree robots. There's a lot of skepticism around the American robotics companies being maybe behind or maybe teleoperating a lot. What's your take right now? Now in humanoid robotics?

Yeah. Specifically, does the US need to pay more attention to unitree running the sort of DJI playbook with humanoids?

I mean, I, I don't know about humanoids man. I mean like, like, like it is, I think it is obvious. It is pretty like we did two features on the Chinese humanoid robotics industry and then Chinese industrial robotics industry and I think the sort of the big markets in the, you know, three year horizon yet say, let's say are much more on the industrial robotics side. But in general like we don't build robots here. And so from a, from a, yeah, I think any sort of like large scale manufacturing thing whether it's unitree or another other or industrial robots or cars or what have you, like China has a really remarkable advantage in scale and manufacturing scale. And the US like it's not just the cost of labor, it's the experience, it's the network and it's the kind of like 25 years of learning that all these firms have been doing to get to the place where they can manufacture a drone 15 times cheaper than the US can. So, so yeah, I think it's, I think it's real. I don't really know how to solve it. I mean you're friends with all the Gundo Bros, like ask them for the, for what they need to build a billion of these. But yeah.

What's your take on the news out of Ant? Talking about how they've had some training model breakthroughs through like leveraging what they're saying is entirely Chinese chips. Do you, do you have a good read on, on that situation yet? Is it, is it important or is it just another headline?

I don't really buy it yet. I think there is a, there's a really interesting wrinkle in the sort of Chinese domestic chip manufacturing arc. So just to back up for all the viewers out there, America in October of 2020 or by the Biden administration in October of 2022, to have this big export control push where they were just like we're going to do everything in our power to. Well, we're going to start trying to restrict the China's ability to import semiconductor manufacturing equipment so that they would not be able to, to make frontier AI chips to train the next generation of models and kind of ever since Huawei, the China's leading chip designer and smic, the kind of analog to tsmc have been trying to push back against that, you know, fight through loopholes and make the sort of level of chips and the quantity of sort of the quantity, the quantity and quality of chips that that Nvidia is able to do at tsmc. So you know, it is a big open question whether or not they'll get there. I think the jury is very much still out. I would be kind of wary of headlines because was the, the sort of most important. There are two important facts to understand looking at this over the next three years. First, Huawei was able to manufacture an enormous amount of chips at TSMC by basically creating a shell company tsmc, you know, deciding to look the other way about some Chinese firm manufacturing all these AI chips and then the US government catching them and be like, what the fuck, you can't do this anymore. So um, they have an enormous amount of supply and so any sort of we train this on only Chinese AI chips is not actually like SMIC chips. It's sort of like, you know, off brand TSMC chips. So the big challenge is going to be whether SMIC can do them, can do kind of competitive chips at scale domestically. And the challenge there is they are not allowed cloud and have not yet been able to replace electrolyth EUV tools which is what ASML makes. And you can't really sneak it in. It's incredibly difficult to sort of re engineer and that's really the final frontier for Chinese domestication. And despite a handful of headlines, I think that over the past week I think that is much more smoke and fire.

So I mean they do have a competitor to as USML in SME. Correct. And then they also need to rebuild SK Hynix at some point I imagine for the, for the memory in the flash.

Right?

Yeah, I mean they were also able to stockpile an enormous amount. It was so awkward because there is literally a Reuters article in like the summer of 2024. Like BIS is planning to crack down on memory and then they did it a little bit but like wrong in October and then they finally didn't fix it until like the, like, like a week before the Trump administration came in. So just fuck ups all around and yeah, at some point, but there's, there's, there's a whole lot of memory sitting, sitting in warehouses in China that they'll be good through for at least the next two years.

All right, so I'm going to massively generalize here and then you can try to piece it apart and figure out if there's any meaning here. But China had, you know, decades to sort of embed Chinese, either former or current Chinese nationals in US companies which then were able to over time bring back sort of important information IP in different ways to sort of like catch up on, you know, advanced, you know, basically catch up on developing their own versions of products from everything from like the F35 to phones and things like that. Right now, China, now we're in a position where like, China is much more advanced in sort of manufacturing, robotics, some of these things that you were saying. And we don't have the same benefit of being able to send a bunch of Americans over there for decades to sort of then like help us re engineer that. What's the US Is like actually viable strategy to kind of like catch up again. China's caught up on product development. Can we then can we catch back up on advanced manufacturing? And what would be. Yeah, is it, is it possible? Right, yeah.

So the thing that I always used to hold my hat on was America attracts the best scientists and funds the most science in the world. And that is a thing that may just stop happening because the Trump administration doesn't care about the National Science Foundation. National Institute of Health wants to blow up universities for better or for worse from their perspective. But like, this is the thing that won us the Cold War, is getting the best immigrants and having them like do crazy STEM stuff. And so, yeah, I am worried about this because that was kind of my ace in the hole is like, yes, you know, there will be this sort of like technology flow or sort of like human, human talent flow back and forth between the US And China. I think it's kind of like inhuman almost to cut that off. But the sort of hope and expectation is that like, America is just a better place to live and folks will want to see, stay here and more like America will gain more from that exchange in the long run. Just like what I was talking about in the, in the sort of deep sea context. And when you look at, you know, a lot of the founders of these AI firms, a lot of found a lot of sort of the top research engineers, like an enormous amount, like a. I would say over 75% of them were not born in the US so that is like our real superpower here is this being a country that is attractive to and like, to a certain extent welcomes the world's best talent. And kind of giving that away just makes it a lot more difficult because you do need to run faster on all these different dimensions. And the way you do that, I mean, I buy into this sort of Silicon Valley mindset that there are such things as 10x engineers and you want to be able to capture as much of them as the extraordinary founders or whatever. And sort of losing out on that is. Is going to be, is going to make it a whole lot more tricky.

Last question.

Peter Zeihan, very popular in tech. He likes to talk about how China's, you know, population is in free fall and the Chinese state as we know it is is unsustainable. Peter Zeihan is one of those guys. The criticism is that when you hear him talk about something you know nothing about, you're like, this guy knows everything. He's, like, completely brilliant. And then when you hear him talk about something you actually know about, you're like, what is this guy talking about? I'm sure when you listen to Zeihan talk about China, you have some thoughts. But talk, talk about, you know, he basically is writing them off. He's saying, you know, yes, they're a force, but he's sort of like writing them off long term in many ways due to the demographic issues, you know. Do you have any comments on that?

What's that Mad Men line? I don't think about you at all. Like, I think. I think in general, sort of there are nuances to everything I've said here, which I have, you know, kind of tuned up for, for our new generation of sports call and radio. I think there are demographic challenges. China is not 10,000ft tall. There are definitely things that it has been really overperforming on. And some of the trend lines. Lines, you know, some of the trend lines, I think are very worry, worry worrisome to Washington. Other trend lines are very worrisome. If you're running China and kind of understanding the nuances of that and, and also baking in, like, different potential futures of, like, things that America could screw up, things that China could screw up, things that America could screw up. Not relation indirectly to China, but relation to the way it deals with the rest of the world.

World are all.

I guess I got to make the plug now. The sorts of things that we explore on China Talk, our podcast.

We love plugs here. Anyways, for.

For.

For more on that, that's, I guarantee you, more thoughtful and engaging than Peter Zion. Please search China Talk.

One word and your favorite media is the website. Go subscribe. Add it to your podcast player.

Well, you are now our official Eastern correspondent.

We'd love to have you back on. I have like 25 more questions. Energy. We could go through chips more. There's so much we could do. We'd love to have you back. Thanks so much for coming on. This is fantastic.

Talk soon, guys.

Thanks for coming on. Cheers.

And we got some breaking news coming up. A big fundraising round. Over $20 million pouring into friend of the show, Pavel Ospar, Ruha's new startup. He should be joining in just A minute.

Minute.

We're excited to have him on the show to break it down for us. I as soon as I saw the news break on X, I texted Delian three red alert emojis saying get in this thing massive. And I think Delian's hopping in as well.

No way.

Let's go. The big brother come on to pop the little bros bag and tell him.

To clean his camera.

Yeah, clean up the camera.

Got a good view.

You got the both the spare house. This is the first time we're ever making an appearance on anything together as far as, you know, fantastic.

Not the last.

Welcome to the show, guys. Break it down. What's happening? What's the company? How much you raise? What are you doing with the money?

So the company, basically the root of it is healthcare providers in the United States spend a ton of time just dealing with the patients. Paperwork from insurance companies and it really both distracts from patient care. So clinicians are just able to serve less patients. And when something goes wrong clerically, it can actually lead to like a delay in patient care. So depending on like some of our customers, like our autism therapists, and it's like that delay in care for some of those kids can really lead to adverse outcomes. So we built is basically AI that merges basically like moves data from point A to point B. Because fundamentally it's just as much as, you know, maybe other curly hair technologists like to say it's not, it's not as much of a systemic issue as it is a technological issue. And the we're just really like insurers set up reasonable processes and have an appropriate check and balance in this workflow where you as an employer don't want to be paying like unreasonable premium premiums. You as a patient don't want to be getting care that's not medically necessary. But the problem is is like there's a competitive space between all these insurers. There's like 10 different insurers that a individual clinician might work with. So there's 10 different processes they have to manage. So it's really about plugging these two parties, making them just like work together better. We sort of raised $27 million across our seed in series A, 5 million seed, 22 million series a day. And we were taking that capital is we really just want to double down on our core product set. It's basically like expanding into more specialties. And really we want to go a layer deeper into the extent that we help clinical staff today. We really just help with like the Clerical parts of things. But as we sort of sit between these two parties, we really start to understand insurer guidelines. Like we understand why often I reject something and we just make sure that the insurer is able to get get clean, streamlined data that has passed all the basic validation. So that's where we're bringing the capital in, hiring engineers, sales, operations, just really trying to scale the machine and expand on our core product set.

So Bain Capital, did the deal, did the deal get done at Fogo de Chao? You gotta let me know the deal.

Or. Oh, D. Oh, that was D. Question, go ahead.

No, no, that's for you because I saw you posting about it like two years ago.

No, the, you know, I. It was part of the close, more so than anything else.

That's what put you over the, over the finish line. That's great.

Yeah, right at the finish line. No deal. We were fortunate enough where this actually, this deal didn't hit the market. We were fortunate to be able to close it from our New York office, so.

Which, you know, should probably kind of covered. But I think for the broader audience, you know, that are maybe not super familiar with healthcare care, the way that providers have gotten paid has just kind of fundamentally changed over the past like 10 years where basically 10 years ago provider provided care and then went to insurance companies and said, hey, here's basically, you know, sort of what I did, I'd like to get paid, etc. And sometimes the insurance companies would look and be like, whoa, this is like really out of whack. Like this is not what I wanted to pay for. And so these kind of like, you know, sort of misaligned incentives. And so policyhold companies focus on this like new workflow, on basically what's on a prior authorization. So it's been obviously like now you have to go to the insurance company first. But because it's sort of like net new workflow and it slows down care, right? Because before you just go to the doctor, they'll give you the care and then you deal payments. Now the insurance company is set up in the pre work but then because that if that takes too long now all of a sudden you're like slowing.

Down the patient experience, which is also directionally like good.

You don't want to be hit it with like an unexpected medical bill. Right? Like that's a lot of the times when you hear some of these like unexpected medical bills happening is because you don't want the check happening after the fact, you do want it happening upfront. But like with our Technology, you're able to do that upfront check without necessarily delaying the patient getting care.

Yeah, if it takes like 30 days.

And this person needs like some cancer treatment.

Right.

Like you want to be able to like, you know, get this stuff like approved quickly. And so there have been like, with the rollout of this new process, there's definitely been some patient backlash being like, what the hell, like the doctor gave me my care plan, I would like to proceed. And I'm like waiting on my insurer to prove my ability to even go get, you know, care in the first place. Place, you know, before the bill. And so it being fast is like super critical. The other thing that I wanted to mention is, you know, so Pavel co founded of these two guys, Sagar, who was previously a true work, but then the CEO is this guy Jeff Morelli. There was actually my high school buddy that, you know, has known Pavel since he was, you know, sort of 14 years old. You know, the, you know, sort of first time they met I think was like up on the ski slopes in sort of Utah where I peer pressure, smoking some weed for the first time when he was weed.

But allegedly.

Allegedly.

Allegedly.

Allegedly. Officially, according to regulators, Pavel has never smoked marijuana. But you know, I think that provided a good bonding moment. And then Jeff actually came and worked with me on my first company, which was not in the prior authorization space like financial services, but it was actually in the healthcare space.

That's right.

Software for autism therapists. And so this pre existing understanding of like the field, you know, autism therapists, what their workflows were. Again we were more focused on like clinical workflow software. But it all kind of ties together of like, you know, Jeff and I used to work together and you know, he'd been friends with high school. He had known Pavel since he was, you know, sort of 14. Pavel cracked it at ramp and like you know, being in the first 25 employees and helping build out their like, you know, sort of bill pay product which has been phenomenally successful for them. And so actually both were in like this like co founder dating process like whatever. It was like you know, sort of two years ago. And I was like, guys like you guys should 100% consider, you know, sort of, you know, working together given the overlap of interest. And so in some ways what they're working on today is like the perfect marrying of those two backgrounds of, of like, you know, financial, you know, sort of services and what Pavel has done and then just background in like, you know, sort of go to market in healthcare, marry the Two prior auth is like this crazy background trend and you know the Asper ah magic can you know make space factories and it can make patients lives in clinicians lives way easier. Part 2 Hopefully Silna ends up way better than Nightingale ever was.

I love it.

Talk about. So we had Lulu on yesterday. She was talking about the golden ratio ship to. Yeah, I think it was had to have been a pretty intentional decision to wait and announce the company and two separate financings at the same time. Pavel, you're a generational poster, you know, you know, potentially poster of the year if you really get back into the game.

Yeah.

Maybe talk about that decision to not, you know, be posting 10 times a day about the company when there's probably an argument that maybe that would have helped in different ways. But clearly you made an intentional decision decision to just keep quiet and focus on building.

Yeah, I think if it had meaningfully blocked the business anyway to not be talking about it, then we absolutely would have launched. But I was just like, you know, early hiring is predominantly in network and doctors aren't on Twitter. So it wasn't necessarily that we weren't like talking about what we were doing, but it was, we weren't talking about it in front of like a tech audience. I think also just there's a degree of focus gifts that can come from downstream not being in the public eye. I think it's limiting in a lot of capacities. I think as we're scaling. That was from the decision now and I think part of it is just, you know, I think we had a sort of an opportunity when we closed the egg earlier than we anticipated to really just come out and like, you know, I'd say like a bunch of things get lost in especially in like a hype cycles. A lot of things get lost in the noise where it's like even like healthcare AI, it's like is this even real what's going on here? But it was, I think it was really compelling to us to be able to come out of the door and be like, hey, we operate in like 45 plus states, we work with hundreds of insurers, we have tens of thousands of patients. It felt like that was more powerful. And I think it's sort of like, I think it's really important to me that like I work on something very tangible with like less. I didn't want, I don't want to be a hype guy. I hype my stuff up but I want to type it off the back of something like very real. And I think we're coming today with very real business progress to show the world.

Has your hiring criteria changed? You're on record saying that you two. Hiring criteria. You need to have that dog in you and you need to be nice with it. How has that evolved recently?

It's a perfect framework. I don't know why you, why would.

You unpack it a little bit for us? Break it down.

What does it mean to have that dog in you? We sort of sit someone down around. We, we just dig through every life decision they've ever made. And then after that, we, we sit all in a room and we go through the two avenues of like, okay, let's all start. Everyone go around. Do you think they have the dog in them? What's the, what's. What did we learn about them that showed that they have the dog in them? And then are they nice with it? That's just, you know, we sit down, you know, we hire everyone that they're supposed to be nice with something. We don't hire like a ton of like, you know, we don't have a lot of generalist strategists around. It's like very like have a specific skill set and so you're making sure the process sort of shows. Like, I think it goes back to like that tangibility is like, don't get me wrong, we do have generalist work to be done, but you always hire someone like on the back of some specific tangible skill set that we can track for.

That's great.

That makes sense. Talk about the moment you were working on. Integrating GPT3 into like, what you were working on at Ramp was like, was that like, like mind blowing at the time? Did you feel like you had discovered something that, you know, discovered a broader opportunity and you just said, I need to dedicate my life to creating business efficiency using artificial intelligence? I guess just like, talk about that. Talk about that moment.

I think there's a few things I think like, it was Jeff, my co founder. His just like family is on, like, his mom's a nurse, his uncle's the CEO of a hospital system. So just had this like really native understanding of what the problem space was, I think at the time. So we were basically deploying LLMs in the context of, of like pulling out vendor contract data. And I think what I found in my ramp is like, top tier company, best of the best. I think when I was sort of looking at my domain and like just the area of fintech, if you go, if you go talk to like a finance person about their, like, Technology stack, they're pretty happy with it. And so part of what it was, I wanted to be able to operate in a space where the technology we were delivering was truly like revolutionary in nature. Something like 10x but better, not 10% better. So I think healthcare was this like awesome moment where it was like, hey, healthcare data is fundamentally textural in nature. Like, you can't tell me what's wrong with your knee with a bunch of like codes. You need to just like verbally explain to me what's wrong with your knee. And so the sort of combination of all these things is like, actually LLMs are basically able to unlock that 10x experience. And there's the funny thing that happened is that like the last big technological wave, like web apps, which is sort of like web apps and like SAS tools and APIs, which is what a lot of the successful startups of like the late 2010s were. It didn't hit healthcare in the same way because frankly, like a web app isn't like for a physical therapist. A web app isn't like 10x better than like a filing cabinet because you're already in person, there's already these physical like notes. And what you're finding with LLMs is you're able to really drive like both technological waves in. So you're able to do like, we're delivering a ton of value with the same sort of technical skills that I learned in developer. And on top of just like this AI blend is really able. Like you can see some of the quotes on the selma health.com of like, we've just become like, immediately this already artery and like game changer for a lot of our customers, which for me is just like really engaging. Gets me excited to get up and work on the problem every day.

One of the frameworks for looking at these startup ideas is find an industry that's highly fragmented and low nps. Can you talk about the previous structure of the industry prior to you launching?

Yeah, well, what I'll say is that's a great structure if you're an analytical guy, not a vibes guy. So we just like to kind of like the problem. People would talk to us for a long time about it. So that's how we landed on it.

Okay.

Yeah, totally. It is like a real. It is a real fragmented market where frankly there was not meaningful solutions on the market that could solve this in a holistic way. There were little tools here and there that you basically had to. In this, the fragmentation maybe is a little different. Different in that the Tools were all fragmented and you would still have to basically employ someone to go through and use those tools and run the workflows. Where we operate is we're just like an end to end solution. You plug us in. There's no tangential tools. You're. We're just able to get it. So like patient comes in, you give us their insurance card, we tell you exactly what clinical documentation we need and tell you like hey, yep, patient cleared for care. Get them scheduled without having anyone on their end necessarily there to manage the process.

Process.

So I think that's how we think about a lot of the consolidation that we're able to do.

Do you have any advice for founders out there that are seeing your announcement today? All the money you've raised and want to copy.

Go to Fogo to chow.

Yeah.

Post more on Twitter. Generic advice. Is that like or not generic advice? My advice if I had to give the world is. I think Keith has said it but I'll sort of reiterate as someone that's lived it is like work at a great company before starting your own. I think that has been like massively helpful from like basically like inception, we had like seed funding. It allowed us just to take a longer time horizon. How you're thinking about it, it just really shifts the model. When you're talking to candidates, you're. I'm able to point like hey, here's the track record. And I think that's something I reflect on a lot is like that Keith advice. Go work at a ramp. Was a great company, had an opportunity to work there and I think it's just made the first like year and a half of entrepreneurship a lot easier than it would have been other otherwise.

Well, hopefully you provide that for the younger versions of you out there.

We got some young, we got new Pavels.

Yeah. Fantastic.

Last, last question for you. Is there a certain milestone that you want to hit before you cut your hair? Is it 100 million? ARR. Is it 500? Is it a billion?

You just, it might just be the.

Look forever growing it down to your waist and beyond.

I think it aligns like look, it aligns with. Whenever I can start tweeting whatever I want.

Okay, there we go.

There we go. I want to see the hardcore buzz cut that's going to be.

It's primarily like you want some pretty quickly. When they see me to be like, ah, that guy works with computers.

Yeah, everything's computer now. It's great. It's great having you on. You're our new healthcare expert.

Fantastic. We'll have you back soon.

Thank you for having me.

Congratulations.

No, it's been. It's been.

Remember?

I remember I came over to Will's office. It was just you sitting there by yourself. You told me roughly what you were working on. It's fantastic to see all the progress. And I'm. Yeah, thank you. As Joe, as Joe Rogan would say, it's an honor to, you know, cover your fundraising announcement.

Yeah, exactly.

Thanks for having me on, guys. Really appreciate it.

Cheers.

Bye and see you, Delian, later. Couldn't get a word in edgewise, but.

It'S so funny to be like, yeah, little bro, I'm coming on your fundraising announcement. I love it.

That's the nature of the call in show. He's just in the text message group. Can I hop in just to hype up Pavel?

There we go. No, I mean I can't. I can't. Pavel's smart enough to know that he should leverage every single advantage that you have in life. And having a big brother like Delian who can be your hype man can help you avoid pitfalls, help you make sure you're actually focused on.

I mean there's another thing.

Bad investors, et cetera.

I mean there's another thing that, that he's really good at leveraging. And that's eight sleep nights that fuel your best days. Turn any bed into the ultimate sleeping experience. Go to T. Go to 8sleep.com tvpn$350.

Off your pod that fuels your day. I saw Andrew posted 100 sleep score.

Oh no. Devastating. I don't think I'm anywhere near there. I woke up at like 4:00am unintentionally.

I actually, I actually did put up 100 last did. She's crazy.

Oh, you prep this.

Okay, so 43. We have somebody joining the show. We have. Exactly. And don't, don't say any names. This is somebody named Carried no interest. They are a prolific poster on X.

You're gonna hear their voice. You won't see their face.

Quite the following. And he's gonna come on and talk about the, the, the situation with all these AI SDRs, all these companies saying they're going to automate outbound.

And then.

Very exciting. He's gonna dox himself in like two weeks live on the show.

Live on the show.

So I'm very excited for that. But let's bring him in.

Reveal. Let's bring him in. He's coming down. Well, in the meantime we can talk about wander. Find your happy place. Find your happy place. Book A wander with inspiring views. Hotel graded minis, dreamy beds, top tier cleaning and 247 concierge service. It's a vacation home, but better folks. Go to wander.com use code tvpn however.

Good you think it is.

Could be better.

It's better.

Could be better. We've taken you on a whirlwind tour of the world today. We've taken you to China.

We got east coast, we got carried.

No better place live. Welcome to the show. Great picture. How you doing by the way?

My hair is wonderful right now too. I want to show it off so bad.

But you sound, I mean, are you doing a Danny DeVito impression right now? You sound kind of like him. You like that?

No, just a lot of smokes boys, that's all.

Okay, I've almost said your name 10 times already but. But anyways for everybody.

This is carried no interest.

He is a private equity investor skewing towards software. Former head of AI at a billion dollar PE fund and doing a bunch of interesting stuff in SAS and looking at businesses as well that are using AI to sort of reinvent themselves. But you were chirping on the timeline earlier today about 11x wanted to just have you on. I don't want to pile on 11x too hard, but I do want to talk about the broader sort of like sales automation space.

What's going on with AI SDRs?

So here's my hot take, right? Like some, some software companies will be AI first and succeed massively. I think that the notion of the AI SDR is flawed permanently. Whether it's 11x, whether it's some of the ones from Y Combinator and there's a few different reasons. Like the first reason to me is that cold email is functionally like an alpha gate game. You are constantly trying to take advantage of a bunch of different quirks about cold email so that you can send 5 to 10,000 a day. And when you abstract all of these quirks away, I know there's a bunch of cold email wizards that will agree with me. You actually lose some of that alpha when you just say, hey, here's this tech company. I'm going to let you run large scale cold email. Very tricky, right? Very hard. That's the first piece that I just don't love love. The other piece is like when you think about, when you think about it. To me, there's five reasons that make it tricky. The first reason with these AI SDRs, you're relying on someone else's cold email data. You can guess what happens to your gross margin when that occurs.

Right.

So if you're an AI SDR company, you have this big database of emails, you're mooching off somebody, you're going to take a hit on gross margin just there.

Right.

Right away.

Right.

The second issue is that in my opinion you have a high probability of potentially being embarrassed. So.

And embarrassing the end customer. Right.

Both.

Right.

You have this AI that is sending out emails. Who knows who it targets. You know, that's a tricky situation.

Right.

When you think about the AI utility relative to the magnitude of mistake. Coogan's law. I have to find a way to coin this. I really do. When you think about that ratio, let's think about Cursor really quickly.

Right.

You get all of this instant efficiency from cursor. You get more code written if you know anything about developer environments. If you push bad code, it should get caught in testing. The magnitude of the mistake relative to the efficiency gain of the AI is very low. You get all this output, the mistake is caught before it goes to prod. Win, win for everybody.

Right.

On the AISDR front. No such thing.

Right.

We're doing it live.

Yeah, we talked about this. It's a good point.

There's an opportunity for a high degree of embarrassment and I think that that is actually leading to a lot of churn. So I've seen behind the scenes on a few of these businesses, the churn does not resemble top enterprise software companies. It simply doesn't.

Well, yeah, and one of the issues is you wouldn't want to use an AI SDR on a very important account account. Because let's say you're trying to close an account that could be worth $2 million a year. Well, is it worth any potential embarrassment to just spam them with a bunch of emails where it's like, hey, I saw you live in New York. Have you checked out Central Park? You should see a Broadway show. Right? Like it's like, it's just not worth it. It's like, hey, this could be. Yeah, you know, just put a human creative or at least keep a human in the loop. Yeah, yeah.

And I don't think it's an 11x problem by the way. Way I think it's a entire notion of the idea problem. Here's another tricky part. Let's talk about, and us three have talked about this privately, but you only grow to $100 million in ARR the way cursor did by being self serve. Right. And some of these AI first companies are incredibly self serve, meaning your CAC is extremely low, your onboarding time is very low, you can grow really quickly. That's a double edged sword for a bunch of different reasons. Right. But now let's think about the aisdr. Not only do you have this high potential for a mistake, the onboarding is a whole thing. You need like a full customer success team. You need constant check ins. The amount of opex you're dedicating to maintaining an existing customer with a bad churn rate. That is not the profile of an excellent software business. Does that make sense?

Yeah, yeah. I mean overall I feel like the risk to what AI is doing to these high growth startups is that we're seeing conversion rates go up, arrs skyrocket, but churn rate also goes up. And that's the real underlying question here because we have so few months of churn data on this new generation of companies that look very different and have different switching costs and different installation costs and there's a new hot model every two weeks and so people are bouncing.

Around a lot and so there's like three buckets. To me as like a very boring software investor, you have your very high self serve, very sticky, low ACV and a low cac that is cursor if I remember correctly. I don't know if they spent a dime on marketing, right.

Their AC couple weeks ago they said they'd never spent a dime on marketing. But at some point that will change I'm sure.

But yes, yeah, and so think about that. You're high self serve, you have no customer success, you're very sticky key, your, your contract sizes aren't big but it doesn't matter because your CAC is low. Yep, that's 100 million of ARR in two years. Right now let's think about a business like sales like Salesforce, non self serve, extremely high ACV and very sticky. Also a good software business. Now let's think about an AI SDR software company. No self serve, many touch points, worse gross margin because you have LLM calls and you're depending on somebody else's email and contact data base. High CAC because again you don't have this amazing self serve option. And as we all know now, bad.

Churn, that's not what is. What is in your view the future of enterprise sales broadly? Is it, is it just back to basics, golf courses, car clubs, you know, you're just you know, broing down and you, you know, you meet, you, you, you just become boys with the buyer.

Le Mans, the best deals in software. We done at le Mans and F1 and all these different places.

I think there's no substitute for that. Jordy. For a million dollar contract, there's no substitute. What I do want to highlight is on the notion of the AI SDR. The idea of LLMs automatically market mapping and targeting your Personas but not sending the emails is obviously high utility.

Right?

Like that is a good thing. So I think that the future of like really high ticket software sales is probably LLM assisted market mapping and like Persona identification. Right. And then the golf course.

Right.

I love it.

The only problem is that first step. That's not a venture backable company, gentlemen. That's just a really good feature of ZoomInfo or Apollo or any new market entrant.

What do you think about the cursor model for SDRs? Something I remember there was a company called Streak that was a CRM that plugged into Gmail and it was very self serve. And the idea was you're a small company, maybe you've just been tasked with sales. You don't even have budget from your boss. You just want to speed things up. You plug into Streak in your Gmail account and all of a sudden you can do some mail merge and some automation. Cursor for SDRS might look like some email generation functionality but. But it's very much that Centaur model where the humans working alongside the AI. Could that be the next 100 million ARR company in the next few months?

Well, yeah, and to be clear that the sort of like AI sales copilot probably has 50 companies.

I'm sure.

Yeah, but I'm curious.

No, I think we are, I think we're actually the first ones to ever think of that. Yes, let's incorporate right now.

You know I like to think about, I liked about, I like to think about, about this. Coogan's Law. This also needs to be coined. I don't know what to call it yet. For an AI product, how many times per day does it call an LLM and derive utility? So let's think about cursor. You're constantly coding, maybe three hours of deep coding work a day. You're hitting an LLM API constantly or at least once per 10 minutes. Let's just say great AI first software sticky. Let's go back to what you said said with the notion of an AI SDR self serve. To me the utility is how many times are you really hitting an LLM per hour? Right. I don't know.

Right.

And so to me I think that there is going to be A self serve AI SDR feature. That is nice. Happenstance from Y Combinator did catch my eye. I don't know if you guys saw that. No great tweet about it. It caught my eye. Doesn't seem like $100 million ARR business to me yet.

Right, we'll see. Good luck to them.

You know, I think happenstance is very cool. I don't see it being $100 million of ARR this year or next year. Right.

Gauntlet's been thrown. Laughingstain. You got to do it. You got to hit 100 million and come on the show to prove carry no interest.

Wrong.

Last question. Our mutual friend Jeremy Giffon likes to talk about the sort of iron law of the business universe, which is like if you grow revenue just shockingly quickly, eventually, you know, you might fall back to earth or you could potentially lose it just as quickly. So what's your take on, you know, generally on some of these various, the cursors, the wind surfs, etc. Do you think that revenue, do you think they can get to a point where they sort of have a durable moat or are they going to just be forever relegated to, you know, extreme competition?

I get a lot of, I have a lot of VCs who I talk to in my network that ask me that all the time.

Right.

Here's the real question and I think let's just cut straight to it. If cursor hits 300 or $400 million of revenue, could they IPO and would the share price be supported? Right. Let's cut all the way through it. Right. And I think that the answer is it could go the way of Slack.

Right.

It's a double edged sword and I think Jeremy's right. Wright, whom I love, shout out to Jeremy. I think he's completely right. It's a double edged sword. There's no way around it, right. That as soon as you gain that person as a customer, you could just as easily lose them. I think that there could be a Slack teams situation that plays out with Cursor.

Right.

That classic Slack is amazing. Slack is the trailblazer. And all of a sudden everybody realizes teams is just fine. Right? And Microsoft just decides, you know, it's time to come for them. I think the same issue could happen with Cursor on a variety of dimensions. Yeah. I can't say with any confidence that Cursor's gonna ipo. I do think it could get acquired for a like gangbusters deal. Like insane, right? I don't think it's an IPO worthy company. Given the churn rate and the potential for Microsoft or any of the big kind of like tech distribution companies to to go at them, I think that it's a double edged sword. I could eat all those words. They could ibo next year. What do I know, right?

We'll see.

We'll see. We got to get on with Taipei. It's been fantastic having you close out the show with us. Let let the audience know to go. Give us five stars on iTunes and Spotify. Not itunes. Apple podcasts.

Apple podcasts.

Five stars.

Five stars for these five stars.

Thank you.

Carrie said it. Thank you for joining the show. This is fantastic.

We'll see you tomorrow, everybody.

Take care.

Bye.

The man.