Amazon Earnings, Saquon is a Superhero, Iraqi's Deserve Sesame Street, Real Founders Use Catheters

27:05I said, considered, and it's still got 14 ,000 likes. I was like, how does that even work? This is more organic. It knows it's not slop and LLM would never make that type of mistake. And I said, I said, you got to put a typo in every post to prove that you're human. And AI would never have the creativity to, to think, you know, Oh, I'm going to try to trick people into, to thinking I'm human by making a mistake, right? So anyways, we got some - Fantastic with Saquon Barkley. Tons of great lessons for business there. I mean, the guy, total come from behind story, just absolute grinder, works on all aspects of the game, clearly reviews every piece of footage, really understands how to improve and spends a lot of time in the gym working on the fundamentals.

27:53And that can be true in business. So lots to learn from Saquon on and off the field. I am generally against wearing, you know, jerseys with another dude's name on the back. We're getting some TV jerseys made. That'll be, you know, just Coogan and Hayes, but I'm almost inclined to, you know, try to get a Saquon Jersey by Sunday and just be, you know, just be cheering, you know, not for the game, but for the ad, waiting for that 15 seconds to just, you know, just blow me away. Can't wait. All right, we got some earnings presented by public.com. I'll have public up and you can run through this and I'll source any data we need to kind of support.

28:36Yeah, if you could look up what Amazon's done over the last five days, 24 hours, I want to know. I want the update from you from public.com. But it's earnings season, boys. We covered Google. We covered Apple. We covered Facebook and Meta. And now we're on to Amazon. And I don't think we have this slated. but I mentioned this to you yesterday, Roblox, which, which, uh, that, that, uh, what's that research, uh, Hindenburg research had done a takedown on, they reported a drop in, in sort of usage and like 30 % of the stock got wiped out. So if Hindenburg was still in, uh, in Roblox, they would have done pretty well off of that, but let's get into Amazon stable, humble giant.

29:20Uh, although they are down 3%, um, in the past week. So, I mean, that's a big move for a big company like that. So they just had their earnings. Their shares fell after sales outlook is weaker than expected. The company is expecting to allocate a record amount of capital expenditure to AI. This has been an ongoing trend. Analysts are worried about overspending in the data center world. Some of those investments can be very good, but there is risk because cash is coming out of the balance sheet. And so let's go to the Wall Street Journal. They say, Amazon shares slipped Thursday after the company projected lower than expected sales and operating income and said it plans to allocate a record amount of capital expenditure this year to build artificial intelligence infrastructure.

30:04Amazon is the latest large tech company to unveil big spending plans for AI, even as some investors are starting to question whether the spending spree is prudent. Virtually every application that we know of today is going to be reinvented with AI inside of it. Amazon chief executive Andy Jassy said on a call with analysts. Amazon projected net sales for the first quarter of between$151 billion and$155 billion, as well as operating income between$14 billion and$18 billion. The high end of each outlook came in below Wall Street's expectations. And so the company's net sales last quarter rose 10 % from a year earlier to$187 billion, in line with analysts' expectations.

30:50Net income rose 88 % to$20 billion, higher than predicted. So very good year last year. A little bit rougher going forward. Revenue from Amazon's cloud computing unit, AWS, increased 19 % to$28.79 billion, slightly lower than expected. Analysts said they were closely monitoring this number after disappointing cloud computing results from Alphabet and Microsoft. What you thinking, Jordy? Uh, I'm thinking that Amazon has been weirdly out of the headlines on all things AI CapEx. It's just not something that you're thinking about and not really hearing about much. I mean, obviously people like Dylan Patel are spending a lot of time writing about it, but it hasn't been top of mind for a lot of people.

31:37But given how much of their business is in cloud, you would think that they would be using that to drive the narrative and trying to build momentum around, you know, the narrative, the Amazon narrative around AI. And it just doesn't feel like, it doesn't feel like, you know, they obviously have like the actual like core data center infrastructure and cloud leadership, but it doesn't feel like there's no, they don't, they don't have their like satya, right? Like somebody who's coming out and, you know, going on CNBC, trying to get on Technology Brothers, There's likely things like that. So they just don't feel like they're, while they are like critical and they're running a lot of this, you know, work, they don't feel like they're a big part of the conversation right now.

32:26Well, I think that's a great take. And I think there's mainly two reasons why we're not hearing that. Even though AWS is the oldest cloud computing provider and I think by far the largest. But I think there's two things that are going on. One is that they don't have a direct consumer narrative around AI in the same way. So Google, everyone's saying, oh, how is AI going to affect Google search? You got to tell me about Gemini. They're launching products that directly compete with OpenAI. They have the TPU. They've really talked about this. And the Transformer was invented at Google. So there's a whole bunch of history there around Google and AI.

33:04Everyone kind of expects that. Obviously, Zuck is doing the same thing. He made this massive bet. They've trained Llama. There's a huge narrative there. Zuck's completely driving that. They did the huge AI CapEx, the massive GPU buy. And then even in Microsoft, which is a little bit less focused on AI productization, obviously they're stuffing, you know, co-pilots everywhere, but they have that massive position in open AI. So anytime open AI is in the news, you can talk to Satya. Whereas Amazon, they have deals with Anthropic. They host a lot of stuff. Obviously there's Alexa and there's AI in their products, but it's not as strong of a narrative.

33:40And a lot of it comes from the fact that they're no, they're kind of no longer in founder mode. Like Jeff Bezos is out. And I was trying to understand, um, do you think Jeff Bezos goes back in the next four years? You think he does as CEO? No, no. I just think it's a really interesting, you know, question and conversation to be had. Uh, I bet, I bet he wakes up some days just itching to be back, you know, behind the wheel of, uh, one of the biggest ships in the world. It seems like there's quite a, you know, looking at the core retail business, I think there's lack of potential, like real leadership in AI.

34:23And that's because you think about who goes on CNBC to talk about Amazon's AI prospects. And I can't name that person. And I talk about technology for three hours a day with you here on the show. And so that's probably a problem. And it just shows there's not, you know, we've seen this with Satya in many ways. He's going founder mode, even though he didn't start the company, right? Like he's clearly the face and the leader of that company. And if he decided to leave, stock would drop, you know, I'm sure, you know, in a meaningful way. Then you have Zuck obviously doing his thing. You know, the other issue is the entire retail side of the business.

35:02I actually pulled up a post from Sean Frank, who we've had on the show probably a hundred times now. he has been talking for a very long time about how little new interest there is in selling on Amazon. You remember there was a time when everybody was selling courses on how you can make $5 ,000 a day selling on Amazon. You don't even need your own products. And so Amazon has gotten optimized and optimized and optimized, and they've just kept eating into the profits of the sellers on their platform. It's basically completely pay to play now. It's really hard to drive sales without spending a huge amount of your sales of gross revenue on ads.

35:39So I just pulled this up. There's a post. This was from a week ago, actually. Sean says, why is interest in Amazon businesses collapsed? One, the overall e-com market has collapsed. Almost no deals are being done, and those deals are 50 % of what they would trade for in 2021. Solo stove was worth$2 billion at one point. It's worth under$100 now. That sums up investor interest. Two, sales are stagnant. Amazon marketplace sales are down single digits. That means the average merchant isn't growing all that much. No growth story. So there's even less interest. He says three, all the big buyers went bankrupt.

36:17So you remember companies like Thrasio, and then you have like kind of other alternatives like OpenStore that are playing more on the Shopify space. They aren't buying the same number of companies. And then the big thing is the fees have increased so much that even if your revenue has doubled since 2021, your EBITDA is flat or down. So you've doubled revenue, but your business is maybe worth less than what it was a few years ago. And so that is creating, it's definitely seemingly like a lack of real, real leadership, right? Because if they had proper leadership place and the proper narrative in place, if they said, hey, we are going to be a real player in AI, we're going to spend tens of billions of dollars on new infrastructure, then that might be a bull case, right?

37:03That might say, hey, they're going to take one of their core business lines and there's a real world where they multiply it. But we don't have that. And then we don't have on the retail side, all the sellers are upset. No seller on Amazon is saying, I'm so happy I built my entire business on Amazon, right? They're getting competed by Chinese sellers. They're being knocked off. There's a bunch of trademark infringement that happens. I one of my portfolio companies element LMNT, you know, had a copycat product on the platform that was the exact packaging different name, but a carbon copy of the product and Amazon let it run up to, you know, thousands and thousands and thousands of sales.

37:46I think they've dealt with it at this point, but it's a very hostile environment to be a seller. And we're not seeing real leadership on the cloud side. So yeah, I don't know what to think. I think it's still a pretty good user experience on the platform. I think it's a great value exchange as a user from I pay, I don't know how many, a couple hundred dollars a year. And I get a bunch of content and free shipping and that's great. But I think there's trouble on the retail side. And because ultimately, Amazon's, you know, they're not, you know, now they own inventory. But historically, it was a very, you know, inventory light model.

38:30And that's what one of the reasons it made it such a great business. And that's why Timu, it does like a meaningful amount of Amazon's, you know, net income profit from the retail business, despite being a fraction of the size. And so I'm interested to see how it plays out. But if you're Bezos and you're looking at all of this and you say, I have this juggernaut, the thing that is, you know, the thing that he put his entire life into. And I could see him getting tired of yachts and, you know, hanging out with his wife, who seems to be enamored by every celebrity that, you know, they cross paths with, which has to be a lot.

39:11so eventually he might just decide you know what it's it's time to go founder mode i also think that he clearly has more interest in blue origin now and i like the fact that he and elon seem to be more more um positive some now and uh supporting each other which is which is great to see as well so who knows um but uh but yeah it's a great great american company for the most part and i'd like to see them, uh, you know, really, uh, take the reins and capitalize. Yeah. I would love to see Bezos back at the helm. He clearly has an incredible amount of soft power at Amazon still, and, um, probably could step in.

39:53I think what's very interesting. I mean, I'd love to get a poly market up on does Jeff Bezos go back this year and then roll that over to next year. I know poly market doesn't do long dated things, but if you look at the timeline, Jeff Bezos stepped down in 2021. This was just almost immediately once Lena Kahn came in and the Biden administration came in. And you could maybe read that as he was studying the history of Bill Gates at Microsoft. He lived through that. He was in Seattle when it happened. There's actually a ton of overlap between their families, I believe, and investors because of the Silicon Valley business community.

40:29And what happened in the Microsoft case was Bill Gates was the richest man in the world and had a target on his back and became kind of a scapegoat for monopoly power and then dealt with this like really rough monopoly case. And even though it didn't destroy the company or really break up Microsoft in a meaningful way, I mean, they had to kind of disaggregate Internet Explorer from from windows. The company has still been massively successful. It's one of the biggest companies in the world. They got through it. But when you, when you listen to Bill Gates talk about it, he says it was like incredibly distracting and it really ruined the job for him.

41:09And then he spent the next, and then it kind of broke his brain because he spent the next 20 years just trying to kind of rehab his image through all of these crazy philanthropies. And so you could imagine he's still trying to rehab his image. And the most effective thing I've seen him done lately is he, he did an interview uh and was photographed in a in a bottega veneta suit i thought he looked fantastic yeah i thought he looked cool and so more of that and less you know running experimental vaccine programs on uh you know india yeah he's also a big porsche guy really oh yeah huge porsche guy he uh he was he it was maybe like a little shot across the bow to elon but he was always like i'm a take i'm a tycon guy uh you know oh i don't do model s which i think is a miss but i think he also has like uh like a career gt and 959 uh a 918 uh probably he'll get the new one uh and uh it's i've been i've been uh i've been driving a tycon this week while i'm in south carolina turbo s yeah uh it is uh it's a very fun car it absolutely rips um but uh so i can see that Yeah.

42:20And so what I think about Bezos is 2021, Lena Kahn comes in. And remember, Lena Kahn wrote her senior thesis, like her dissertation on Amazon as a monopoly. And Amazon was not thought to be subject to antitrust legislation because of the fact that it acts as an aggregator, as Ben Thompson described, in the sense that competition is only one click away. Their strength comes not from cornering supply. It comes from cornering demand. And what that means is that Amazon, they might be able to raise prices, but they never have. And so it's impossible to, it's really, really difficult at least to prove consumer harm, which has always been the definition of we got to break up this monopoly.

43:11It's one guy owns all the steel or all the railroads. And then because he owns all the steel, he can say, I'm doubling the price of steel and everyone still has to buy. And you can imagine that that's bad. It makes sense why we have antitrust legislation to prevent that type of stuff. In the case with Amazon, it's seller harm. And so sellers are pissed off. Sellers are still able to eke out a business, but their businesses aren't growing in value. And so that's a major, again, it doesn't mean that there's a strong antitrust case to be built there because as a consumer, I'm sitting here and I'm like, well, I paid 200 bucks a month or 200 bucks a year to Amazon.

43:52I get a bunch of content. I get free whole foods delivery. It's a pretty good exchange. So, yeah, we'll see how it plays out. But with that. And so 2021, Bezos steps down. This is like, you know, within days or within a year of Lena Kahn coming in. And then two years later, the FTC files an antitrust lawsuit formally. And so I think of it as the writing was on the wall. The Biden administration is going to be very rough on me. If I duck out now as CEO, there's no scapegoat to go after. and it'll be much harder to rally public support for an antitrust lawsuit because we won't be able to say, look at this multi-billionaire, look at the richest man in the world.

44:38He's extracting value. Everyone needs to support this. And that support, even though FTC lawsuits are not voted on by the American public, there is pressure that happens through various elected representatives to the FTC. And it's a lot easier to go after someone who looks like a, you know, Lex Luthor than Andy Jassy, who's just a guy who no one really knows his name. And he's just there being like, yeah, we're obsessed with cloud computing. And it's actually a very, it's a very hot market, very competitive. And look at Amazon. Amazon, yeah, we might be the biggest in cloud computing, but Google Cloud is huge.

45:18Azure's huge. And Facebook's also investing in data centers in a massive way. And then there's a million other providers. So it's a complete, perfectly competitive market. Yes, we have great margins, but it's not a monopoly. Yeah. Now you have a bunch of new players like Stargate and you can point to them and be like, look, they're investing $500 billion in their own. So we're not venting new players from entering this market. It's still going to be very competitive. And so with Lina Khan out, it kind of creates a space where Bezos could go back in if he wants. And I've always been interested in, I always love when the founders have been able to set up their businesses in a way that they can do everything in the one master bucket.

46:02Zuck is the master of this. He has the super voting shares. And so when he wants to do the metaverse, that goes in the meta bucket. When he wants to do AI, that goes in the meta bucket. Elon's taken a much more decentralized approach, but then he's re-centralized with things like SolarCity getting bought by Tesla. XAI is integrated with X and Tesla has some people over there. So he's kind of creating this caretzu of different companies that all have intertwined stakes. There's a lot of overlapping investors. People actually think about the portfolio of Elon companies as like Elon Inc now, and that's happening.

46:34It's not all one company. It's crazy to think that he could take just his personal positions public independently and create an index on and it would probably trade at a 5X premium to book value, you know, something ridiculous. Yep. But Bezos has always, he's never been able to justify that fully. Like he, I think Amazon did one satellite internet project, but they, they weren't, he wasn't able to go to the Amazon shareholders, even though he has an immense amount of control, he was able to go to them and say, look, we could be making money on retail, but instead we're going to invest all of that in AWS and we're going to have this amazing second business.

47:12He wasn't able to do that a third or fourth time. He could have said, hey, look, we have a lot of delivery trucks. We want to make electric delivery trucks. And so we're going to build electric vehicles in-house. But that's not what he did. Instead, he funded Rivian. And Rivian is kind of his Tesla. And that stock's been up and down. But Amazon built a position. And Bezos helped with that a little bit. But he wasn't able to just do it inside. If Zuck needs trucks, he's going to build them in meta. He's not going to start a second company. And same thing with Blue Origin. Like, like if Zuck wants to build rockets, it'll probably happen within meta because he has so much control and so much voting power there.

47:46And he's, and he's been very good at convincing the shareholders. So it'll be interesting to see. I mean, I would be most, I would, I would honestly probably be more excited about Blue Origin and Rivian if they lived within the Amazon holding structure. And Bezos was at the top of that and could really choose how to allocate capital because this, this business, even though, you know, it's down a couple percent, it's throwing off so much money that they are going to invest. a hundred billion dollars in capital expenditure next year like that's probably way less than rivian needs or blue origin needs these could be within think about how good that would be for users if you could be driving in your rivian and click for to do an amazon haul and buy a 13 dollar uh guitar or 13 dollar couch right 20 dollar couch yeah yeah that's what that's what users want right they want slop products while they're in their you know slop ev yeah yeah i I mean, at the same time, it makes sense.

48:40My shareholders are like, wait, like I wanted to like Zuck, I'm here as a shareholder. I just want like ad dividends. Like, why are you open sourcing LLMs? Like, I don't care about that. And Zuck has to make the argument that, oh, well, it'll help with talent and it helps with our, it's commoditizing our compliments and, oh, we'll be able to stuff this and everything and we'll have a stake in AI. And it's obviously gone very well. But even as a CEO who controls a massive public company, you do still have to make your case to shareholders. You can't just say, I have a wild idea. I have no idea how it connects, and I'm going to do it anyway.

49:16People have to say, hey, it's going to make money for the core business. But Amazon, I mean, it really is interesting, your point about them not being like leading the AI narrative, because they are leading the AI narrative. They're spending$100 billion in CapEx this year as part of a push to build generative AI services. They're leading the investment in CapEx. They're not leading the narrative. Yeah. And I think that narrative just matters so much for a company, you know, that, you know, every company is in the game of we want our, you know, for the average, you know, retail investor that owns Amazon.

49:56They want the, you know, they want the price, everybody wants the price to go up. And so I think that they're not getting credit for that CapEx from a market cap standpoint. 100%. And so let's just break down how Amazon compares to the other hyperscalers. It really is fascinating. You know, Satya has that quote, I'm good for my 80 billion. Well, now that has actually increased. Amazon, Microsoft is expected to top 90 billion in CapEx this calendar year. Google is spending 75 billion and Meta is spending 65 billion. And so you add all that up, you're in some huge, huge territory, 350 billion, something like that.

50:39Pretty, pretty crazy numbers. But you know, it's a lot of data centers. I'm excited to see it. And it's the meme, another trillion to one, another trillion to NVIDIA. Woo-hoo. Yeah, we need to do a data center. Yeah, for sure. For sure. And so US tech stocks tumbled last week after the surprise success of a new generative AI model from China. That's of course DeepSeek. The free to use model was built at a fraction of the cost, although very debatable at this point. The latest number I saw was like, they spent 1.5 billion, which is like more than it cost to train GPT-4. And so now we're in like weird territory.

51:22Yeah, they spent$1.5 billion while horrifically stealing from OpenAI in order to, and then, you know, lying about it. I'm actually very confused about the real number here because it started at$6 million. Then people said, well, they have other stuff and it doesn't include R &D. And even DeepSeek admitted like we didn't include all the cost. And so people were like, okay, maybe it's$50 million. Maybe it's$100 million. It's still impressive. They clearly did a bunch of optimizations to make the code run faster. And there's just no question that inference is cheaper. Like that is true. We know that.

51:57But now we're getting these crazy estimates like actually DeepSeek has invested$1.5 billion in AI as a whole project. But then on the other end, you're hearing like, oh, a couple kids at Stanford rebuilt it for$50. And so we're now like in the bands of like DeepSeek spent a billion, but Stanford can do it for$50. I don't know what to believe anymore. It's crazy. I love how they just inspired America to grind harder. Everybody was like, oh, it's possible to train a model like this for very, very little comparable cost. We should just do it. And then they do it. so i love it i love it well uh and so uh obviously that shook markets amazon ceo jassy wants to make amazon an ai leader as the world he needs to do more podcasts though he needs to start setting the narrative call lulu go direct as the world's largest cloud service provider the company is in a good position to profit from ai amazon created special teams uh created special teams to uh let's see, created special teams to drive generative AI innovation and has released a flurry of services, including an AI shopping assistant.

53:08First time I'm learning of that. I've heard that they have an AI chat bot that you can chat with. It's so funny. Their shopping assistant for the consumer is, yeah, we're just going to find you the products that you really want. And then on the backend, they're just telling all the retailers, yeah you're going to have to give up 90 of your margin to get this customer and you're going to love it yeah otherwise you know somebody else will pay for that slot so and and so the the wall street journal highlights uh amazon's competitiveness in the e-commerce business with new rivals such as timu and sheen uh both these companies offer low prices to source products extensively from china uh president trump suspended a trade exemption known as the de minimis provision that sheen and timu and other amazon competitors have used to avoid import duties on low value packages from China.

53:57We've covered this before. And honestly, we never thought that it was putting a dent in Amazon's revenues because a lot of the slop products that were bought on Timu and Shein were incremental and it was seen more as gambling dollars than really replacing or displacing Amazon as a staple of the American consumer's uh uh consumer habits so uh unclear how big of an issue timu and sheen have been historically they're certainly on their back foot now and it seems like amazon's even even better positioned i mean a lot of this although the stock traded down it there's a lot of bullish stuff in here like they've they've they've they've like there's no more ftc risk with lena khan out uh they they're they're potentially killing off team machine and, and they're spending the most on AI data centers.

54:49So if you believe that having a lot of data centers, a lot of GPUs is valuable. They're in the best position now, not by a lot. They're only spending 10 % more than, than Microsoft. But it's still a lot and they got a lot of stuff. Yes. It's great. And to be clear, the stock is not under performing by any means on earnings, but I'm on public and they are up 5 % ish. in the last month. So, um, not, not a bad spot to be in, but I think they were up 40 % last year, which is crazy for a company that big. Yeah. They were not a, they were not a trillion dollar company in 2020. They were, um, in, yeah.

55:29So right. Uh, Yeah, if you bought Amazon the day that the NBA basically went offline, it was a$900 billion company. right wow um so anyways still doing well despite bezos not at the helm but uh imagine imagine the move if bezos did join the company the the the pop would be uh biblical for sure for sure uh and so separately during the important holiday quarter amazon grappled with threats from one of the largest US labor unions. And I love their name, although I'm not a fan of them personally. The International Brotherhood of Teamsters. I know the Teamsters Union. I didn't know that was their full name.

56:19We should start an International Brotherhood of Technologists. They tried to disrupt holiday deliveries to pressure Amazon to negotiate with its warehouse workers. At the time, the company said those attempts had fallen short, I guess. And so Amazon got through that. I also wanted to go back and look at a little bit of older earnings analysis from Ben Thompson. This is from 2023, talking about Amazon and AI. And Ben Thompson is highlighting the lack of clear narrative. And so Jassy, at the time, back in 2023, he says, generative AI has captured people's imagination, but most people are talking about the application layer, specifically what OpenAI has done with ChatGPT, it's important to remember that we're in the very early days of adoption and the success of generative AI, and that consumer applications are only one layer of opportunity.

57:13And so three of the five big tech companies have a clear AI story at this point. Microsoft has its partnership with OpenAI. Google has its own AI models and Google Cloud Platform. And Meta is pursuing open source strategy in conjunction with seeking partnerships with large cloud providers, starting with Microsoft. That leaves Apple and Amazon, who have may underperformed on the narrative front. Yeah. And so here's the thing. So Amazon made its bet on Anthropic and Anthropic has built a great model. People love Claude, but is kind of irrelevant on the consumer side. They have a fraction of the downloads and the engagement that OpenAI does.

57:52And they don't have a reasoning model right now. They do. It's just internal. They haven't launched it. That's what I'm saying. I don't care. They don't have a, they don't have a, you don't get credit for, oh yeah, we, we got, we got a, we got a great, like the whole thing with, with all these, you know, foundation model, you know, companies is they're launching products oftentimes before they're really ready for production. They launched deep research and you didn't even have access to it. Right. So everybody's sort of running. I believe that the Anthropics model was actually competitive. They would have launched it already.

58:26Yeah. Yeah, it is clearly a very, very different culture. And the other thing is, Anthropic made a clear bet around AI safety. And a lot of other firms ignored that, DeepSeek, one of them. And so that seems to be in the short term the wrong bet to make. And so that was Amazon's real player in the race. And I think that, or at least consumer narrative, and it's turning out that Anthropic doesn't seem to have a strong consumer narrative right now, even though people love the product. Yeah, it is crazy comparing. I haven't personally used Claude in weeks. So yeah, I mean, it seems like it has popularity amongst the developer community, but beyond that, it hasn't really broken through.

59:13And it's fascinating to see how aggressive Sam has been at OpenAI about responding to just any developments. You know, every time Google has a keynote, OpenAI has a keynote two days before. And DeepSeek comes out. What's the name of the product he launches? DeepResearch. He wants the Google autofilled AI Deep something. Oh, yeah, DeepResearch. Go over here. And you can tell that DeepResearch, I mean, I've been enjoying the product and I think it's getting a lot better. But you can tell that they're not ready to scale that for a few reasons. First, it's not available on mobile. So the button is just not available there.

59:53So you have to, if you're on your phone, you have to open up a web browser and go to chatgpt.com and then click the deep research thing to do a query on mobile. So they're clearly cutting the amount of deep research queries that way. And then also, it always asks you a clarifying question to slow you down. And I'm sure there's some sort of churn rate between, oh, I was about to kick off this deep research thing, but it asked me a bunch of clarifying questions. Maybe I'm not going to do it. And if that saves them 50 % of the queries, that's a lot of compute because clearly it's very, very expensive.

1:00:25But it shows that they're willing to push it out and get the data they can and do as much scaling as they can, even though they're not ready for this to be free for everyone, ad supported, and billions of inferences that are just burning up data centers every single day. Anthropic, all of this stuff, DeepSeq coming out and people responding to it, and it's kind of showing that we may be at least a few models away from the sort of asymmetric risk that Anthropic has been focused on. And so it seems to have been so far, at least for these early reasoning models that aren't fully, you know, agentic.

1:01:09It does seem that they kind of made the wrong bet there. Yeah, it'll be it'll be interesting to see how it plays out. Maybe Bezos gets in there and put some pressure on them and put their feet to the fire and says, give me the models. They're going out today. But let's close with Ben Thompson's analysis. this is again from two years ago talking about Amazon's play in AI and he has an anecdote. Ben writes, I'm wrapping up my usual summer stint in the US in a metro area that started receiving same day deliveries earlier this year. Speaking personally, I absolutely agree with Jassy's characterization of consumer behavior.

1:01:45I both found Amazon's delivery speed to be astounding and also recognize a significant shift in my own behavior because of that speed. The truth is that buying something locally in person still requires a trip of some sort with the risk that what you want may not be in stock is in almost all cases, it was actually faster, particularly once you incorporate the realities of life where you can't actually go to the store this very minute because things come up to simply order from Amazon because you'll receive your item in a matter of hours, not days indeed. Indeed, I felt this summer marked the largest shift in my purchasing behavior since the launch of Prime.

1:02:21When I needed something like, say, a mic cable, which happened last Sunday, I simply bought it on Amazon the moment the need arose. The cable arrived four hours later. What I find so fascinating about this, though, is Jassy's argument that this actually results in lower unit costs because there are fewer touch points for the delivery. This does not make intuitive sense in that it is a classic tech trade-off between marginal costs and fixed costs. It is also a tangible expression of why I thought it was so important for Amazon to push for this level of delivery speed in the first place. Amazon is at its best when it's realizing advantages that no one else can match because of its willingness to invest in the real world.

1:02:59This is what Amazon does best. Focus on long-term moats and instead of short-term profitability, that's what leads to increased profitability in the long run. If you're patient enough and confident in your position with consumers will endure, Amazon's renewed interest in its logistics is a perfect example and an encouraging sign that the company will retain its greatness without its earnings call skipping founder. Because Bezos didn't go to the earnings call, which I think is funny. Anyway, Jordy, what you got? Yeah, I would just kind of close this out by saying Amazon generally is very well positioned for a lot of the trends that we're seeing right now.

1:03:38Some of these other companies that are more pure play software companies that aren't existing in the real world, there's some real threats there, right? Software development timelines are changing so dramatically. It's becoming so much faster and cheaper to make software. That threatens a company like Salesforce in a way that Amazon, which is selling physical products, selling content, selling, you know, they have their logistics network. They have their data center business. they they should be people should be tremendously excited about amazon's positioning for the next 10 years and yeah they're somehow not getting that through so anyways excited to keep covering it yeah it's it's an interesting story seems mostly uh narrative and communications based but we'll see fortunately you know stocks up 40 over the last year uh minor minor bump in the road with this earnings.

1:04:31A little bit of scared on the CapEx, a little bit of deep seek threat, but we'll see where it goes. Let's move on to our next top story. We have a Technology Brothers exclusive. For the first time in Technology Brothers history, we're covering an Eric Newcomer exclusive on General Catalyst's fundraising. And so we have slides here from Eric Newcomer showing how General Catalyst has performed, how they've scaled. And Eric Newcomer has the firm's Q3 2023 fundraising documents. And General Catalyst is on a tear. They are just going big, ultimate size lords. Let's hit the size gong a couple times for them.

1:05:11They're just massive. I don't think this article would have caught it, but they came out yesterday announcing they have a new fund of funds platform to fund other emerging managers. So they're just going, I love, we love when people get aggressive and they're certainly doing that. Yeah. I mean, the narrative around General Catalyst has been it was a VC firm, then it was a growth stage firm, then it was a private equity firm. Now it's just an asset manager in every sense of the word. I think they also launched a wealth management product, right? Yeah. Did you see that? Yeah. And so they have a hospital, consumer value financing, a wealth manager under the leadership of CEO Hamant Tanaja.

1:05:55The firm has grand ambitions. Newcomer got his hands on documents that show how General Catalyst has presented itself to prospective investors. Documents he obtained are admittedly a little dated from 2023. That's a bit old, newcomer. Got to step it up. gotta be hanging out at the sandhill uh the sandhill rosewood a little bit more yeah or the nobu roshima hanging out uh yeah i want to see i want to see i want to see a tech journalist go to japan study how to make sushi for 10 years and then go get a job at nobu and just sit at the sushi bar you're just making sushi everybody thinks you're just you know making a nice hand roll turns out you're listening for scoops you're picking up every single thing that they say you're yelling irashima say but then you're really you got your ears both ears here because you got a bunch of allocators right in front of you you know you're picking up this round's happening that round's happening maybe a faster way to do that would just be to go pay off the individual you know the the wait staff the sushi the sushi uh chefs etc um but there's alpha everywhere folks And there's certainly Alpha and the Palo Alto Nobu.

1:07:07Yeah, newcomer, become a janitor at Amman. And then just go around and just sweep up all the slide decks that are laying around the pool and the jacuzzi and the spa. And just scan them, throw them up on the newsletter. You're going to triple subscriptions. You're going to be printing$5 million next year in EBITDA. You're playing small ball right now. Yeah. Yeah. And so the investment memo provides a detailed look at General Catalyst's strategy and performance as it was raising its last suite of venture funds. And so let's go through some of these. They are absolute dogs, General Catalyst. Billions raised in October 2024.

1:07:44The firm announced$8 billion of new funds for its Fund 12. been around for a long time, including$4.5 billion for the firm's core VC funds, $1.5 billion for its creation strategy, and$2 billion for separately managed accounts. $1.5 billion for, I'm reading that as incubations. Is that correct? I think so. $1.5 billion for incubations is insane. That's just an insane number. Given how capital efficient incubations can be, because sometimes a firm will put a founder as an EIR, so they have a bit of an income and benefits while they figure out what to build. I have no idea how it's actually structured.

1:08:29Well, their latest incubation was a roll-up. And so they wound up buying a lot of existing firms. And so you can imagine if you're buying like a mid-sized business, dropping AI on top of it. Yeah, it's going to be a hundred million equity check and then a couple of million, hundred million in debt maybe. That's going to go pretty quickly. That's 15 deals if you're doing a hundred million dollar checks to kick off these like incubation roll-up type things. I have no idea how they're sizing their bets, but I'm sure we'll find out more. Let's go to the evolution. In the year 2000, the firm started in Boston, founded by serial entrepreneurs David Falco and Joel Cutter during the dot-com bubble.

1:09:11General Catalyst established credibility. Let's see. They hatched winning companies, resulting in Kayak, Demandware, and Virtu. Fund 5 in 2007 was$722 million, already pretty big. During the global financial crisis, they expanded to Silicon Valley in 2010 and New York in 2012. They did a fund six of$500 million. Fund seven was$676 million. Every fund is just getting bigger and bigger. Ken Chenault, who I've actually interviewed, ramp investor, former CEO of Amex, he joins the chairman. And then in 2018, General Catalyst solidifies a full stack platform by raising multiple funds segmented by stage.

1:10:00This is the Fund 10 group. Fund 10 comes in at$2.9 billion. General Catalyst raises its first sector-specific fund, the Health Assurance Fund. Hamant becomes CEO. They open a London office in 2021 post-COVID and Fund 11 clocks in at$5.4 billion, baby. And so there's a little bit of a breakdown here for Fund 12. They have their creation fund, which hatches companies, transforming companies with a catalyst advisor and executing venture buyouts. Okay, so that's interesting. So it's their creation fund, but it's a venture buyout strategy. So that makes more sense. Yeah, and they're trying to own 25 % of the business.

1:10:46Trying to deploy it$1.5 billion when you're creating pre-seed companies and presumably making room for other investors to just... Impossible. Difficult. But they're going to buy companies like their private equity firm, but where they think there is VC-like opportunity and venture scale. And so they have the Ignition Fund. This is where they do their core early stage work. Another 1.5 billion there. They focus on Seed Series A, Series B. And of course, they focus in 10 to 25 % ownership target there. They have their health assurance fund, which is targeted at digital transformation of the healthcare industry.

1:11:28750 million early stage seed series A, series B in healthcare, 10 to 25 % ownership targets. And then they have their endurance fund. That's their growth in later stage investments, 2.25 billion. And this is all out of fund 12. And so they've taken this full stack approach. You can look at some of their leadership here. They focus on basically everything, healthcare, fintech, global resilience, technology, consumer, and enterprise. And we can go through more of the slides, but you might just want to subscribe to Newcomer. The guy gets scoops, even if they're a year and a half old. Always scooping.

1:12:07Always scooping. Yeah, an absolute dog. Is there anything else we should talk about with regard to General Catalyst? We got some boys over there, some absolute dogs, writing huge checks, making plays. You know, every time a venture fund comes out and announces multiple billion dollars of new AUM, there's always naysayers, haters saying, oh, this is not real venture capital. Or, you know, how are you going to actually put up returns like this? But you have to just give people a little bit of credit. If somebody is raising$8 billion, unless they're Masa son, they have probably planned things out pretty precisely.

1:12:55They probably have a very coherent strategy because that kind of money doesn't change hands and get invested without, you know, the people that are investing those dollars are sophisticated. And yeah, they're taking a portfolio approach. They don't need every fund they invest in to outperform. But yeah, got to give credit. And I like to see that General Catalyst is getting aggressive and evolving venture. It's not a cottage industry anymore. I think that the super, super early stage stuff will continue to be somewhat boutique. And there's always going to be opportunities for new managers. But, you know, the bifurcation venture from these sort of small 10 to 50 million dollar funds all the way to these like five to 20 billion dollar funds or five to 10 billion dollar funds.

1:13:44That is seemingly the new normal. And I think it's overall healthy. It's important for early stage investors to know that, you know, if I invest in a company at a five to ten million dollar valuation, that if they do well, there's going to be, you know, 50 billion dollars of downstream capital that somebody like a GC or an Andreessen or a Founders Fund or a Sequoia can can deploy into that. So overall, I think it's healthy. I think there's going to be an entirely it's going to be cool to see the generation of of investors, allocators that come out of firms like General Catalyst that were there during the transformation and then can continue to innovate down the road.

1:14:25So overall, very cool. And, you know, look, they're probably doing a few hundred million ARR just on the fee side. Yeah. So the old model of two and 20 is gone. VC firms used to take 2 % annual management fees and 20 % carry, but not a general catalyst, baby. The firm share profits on investments starts at 25 % and climbs to 30 % if the profits are above 133 % in aggregate contributions. And so a lot of these different funds have different management fee structures. The Creation Fund gets 2.5 % management fee. The Ignition Fund gets 2.25%. Endurance is their growth fund, so a little bit lower, 1.75 % management fee.

1:15:12And the Health Assurance Fund gets 2.5 % management fees. So Eric Newcomer clocks it at$120 million a year just from Fund 12. And remember, they're raising these new funds every couple of years. And so that's going to fund a lot of employees, a lot of office space and a lot of ski trips to Tahoe and Sun Valley. Yeah. And one thing to note yesterday, we will probably talk about this later in the show, but yesterday some news broke, I think from Bloomberg that Trump was exploring eliminating what has been called the carried interest tax loophole. And I think that loophole is a sort of not the best way to characterize it because what they're referring to is the fact that investors that are, you know, people that manage and deploy money that's not necessarily their own.

1:16:07they get when they you know, if you have a 10 million dollar fund and it does well and you're in carry that carry gets, you know, has has historically been treated and today gets treated as capital gains. What they're trying to do is make it treated more like ordinary income. So instead of paying a, you know, 15, 20 ish percent tax rate, whatever it is, you might be paying closer to 50 percent tax rate on that. And so for someone like GC, who's deploying billions of dollars and presumably wants to make a great return on that, that's billions of dollars that they would be giving to the federal government or the state government, depending on their performance.

1:16:50So pretty would be pretty brutal for the industry in many ways. But it's coming from somebody like Trump who runs his own money. So as a real estate guy, he's using his money. I'm sure he has co-investors and using a lot of debt, but to him, it maybe is not so important how carried interest is treated. But if you're a fund manager raising billions or if you're a fund manager that's raising 10 million and really making money on the carry piece versus the fee side, it would be pretty devastating. So, yeah, the carried interest loophole closure, this whole discussion that came up during the last administration didn't go anywhere.

1:17:36You said that if it goes through, VCs will make January 6th look like a baby shower. I think we can see lots of VC and private equity associates protesting on the National Mall. If this happens, it'd be very bad for the industry. And who else said something interesting? I forget who it was. Dan Primack. Dan Primack says, a source on the phone just now on the carried interest tax situation is like the red wedding. Trump collected all the VCs at the White House so he could more easily murder them. And so - Yeah. And to be clear, it's not going to be the end of the world, but it does meaningfully impact.

1:18:27When I look at it, it's like from time to time, I'll do an SPV into a company that I'm excited about. And now I have to, if they do change the law and the tax treatment right now, if I invest a million dollars, be an SPV, I take 20 % carry on it. Maybe there's other partners in the deal, so I'm not actually getting the full 20 % carry after it's taxed 50%, it might not have even been worth my time to do the deal. I might've been better off just putting in, you know, 50 grand myself, letting it ride and knowing that it's going to be treated like, um, cause, cause putting together money like that via an SPV, it's a lot of work.

1:19:08It's super time intensive. And so, um, you know, you're basically looking at the way that I've always looked at it is you're investing in the deal with your time. And so from that lens, it's saying, okay, maybe it should be treated like ordinary income, but the payoff is so unclear and it takes so long to see a payout and you only see a payout if there's a meaningful return. And so I see a good argument to keep it as is. I don't think it should be framed as a loophole, but we'll see how it plays out. And so Manifold Markets, I don't know how up-to-date this is, but they are running a market on carried interest repeal by the end of 2026.

1:19:51It's at yes, 40%, no, 60%. I imagine that's pretty low liquidity. Med Faber here, who actually follows me, says, We need a polymarket on the chances of carried interest, LOL. Close the carried interest tax deduction loophole. So everyone wants to know whether or not this will happen. And it could just be one of those things where, like Trump is doing with Canada, oh, we're going to make Canada a state, this and that. Trump loves to open with a crazy proposal and then back it off and get something in trade. And so I could imagine this saying, look, I'm going to close the carried interest loophole, blow up your entire industry.

1:20:29But as this gets, now that I brought everyone to the negotiating table and people realize how serious I am, Trump will say, well, actually, I'm going to close the carried interest loophole. on investments that are made outside of the United States. Or, and really his goal the whole time has been, I just want VCs to invest in American companies that make products in America. And you can see that being more aligned and that's actually what he's going for. But how does he get there? Well, he's gotta come in with a hammer and get everyone waking up and saying, look, it's gonna be the red wedding. There's gonna be a bloodbath unless you play ball.

1:21:03And then, you know, he set the bar here. And so when he backs down, he gets exactly what he wants. and so i could see that happening and maybe that maybe that would be good for america it's possible let art of the deal let art of the deal um okay let's go through some timeline uh and uh ben if there's any questions or reviews you want us to read send those to the chat and i'll cut over to jordy uh as we go through but first we got a promoted post from one of the favorite companies on the channel i'll let you take this jordy those that don't know i'm a co-founder of rora Aurora makes beautiful, easy to use, highly effective water filters.

1:21:42I just thought this was a cool shot. Brian, CEO, posted this yesterday. It's just showing the same exact tap water coming out of NYC pipes. One on the left going through Aurora, one on the right just sitting there. And you can just see the difference. You can see the contamination of the water. So if you're drinking tap water, you've got heavy metals, pharmaceuticals, tons of stuff in it that you just don't want to be drinking. It's absolutely insane to drink tap water. I simply, I haven't, I drank tap water once in literally the last like five years. And it was actually at David Senra's event, not to throw David under the bus or anything, but I got back to my room and it was at a, you know, it was at this really cool retreat center.

1:22:32And I was like, probably like a 20 minute walk from where I could get water. and I didn't even know if it was open. So I drank tap water. I was like, it was, it was in Austin, which has notoriously bad tap water. And I was like, this is egregiously bad. Spit it out and just ended up, you know, it's going to bed thirsty. But, but anyways, filter your water. Aurora makes good filters. Highly recommend you check them out and feel free to DM me if you have any questions on it. Fantastic. And one, I've had multiple calls now with Technology Brothers listeners. where I popped into the call and I already had their Aurora in the background.

1:23:10So very, very awesome moment. Thank you for the support. You'd love to see it. Well, we got some major personnel news in the world of AI. OpenAI co-founder John Shulman is joining Mira Morati's startup after a brief stint at Anthropic. The trade deals, it's like the NFL. The Holy Trinity right here. Yeah, the Holy Trinity. It really is like the NFL. I mean, these guys are bouncing around like crazy. And you never know if it's like, oh, it's AI safety. It was too dangerous. They were just too reckless. Or it was just like, I just didn't like working with those people. Or I wasn't getting comped right.

1:23:50Or, you know, some deal fell through. I wanted a different thing. Or I wanted to be more product focused or more research focused. But everyone's excited about Mira. I mean, look at the pressure that Sam is under. Elon's attacking him. Two of his former co-founders, Ilya and Mira, both have competing startups. He's competing with Anthropic, DeepSeek. It's not so often that, I mean, it happens, right, where you have a group of co-founders that build a big company and then one leaves and starts a competitor. But not in a while, and certainly not in my memory, can I think of a time when a relatively new company that was already a multi-billion dollar company had two co-founders leave that each now have their own billion dollar companies that are generally competing for the same opportunities i think ilia was in the news today this morning that he's raised uh potentially raising new money at a 20 billion dollar valuation and so it's got people are clearly bullish on the open ai founding team and this is a big win for mirror because you can imagine that John could have gone to Ilya's company as well.

1:25:01And so anyways, this is just a great time to be in the technology temple that is our studio and my remote studio. I love it. It's fantastically dramatic. The weird thing that I never understood was like these departures are always framed as like indictments of open AI. and yet it's like okay so you guys all don't like open ai but you also don't like each other because otherwise you'd all just start a company together and you all started separate companies and so clearly like the infighting at these companies is just crazy and in many ways that we like the internal politics are much more complex than just like oh there's two sides to this issue because like why why aren't mira and ilio working together why did john shulman leave philanthropic.

1:25:46Like, like you would think that they would just be like the, the open AI and the anti-open AI, but instead it's like, there's the anti-open AI and then there's a different one. And then this one's more about safety and this one's more about this and product and stuff, but either way, it's going to mean a lot of great AI products and I'm excited to try them all and let's see what happens. Yeah. It's really a crazy time. I mean, when the pitch with AI right now, the reason to invest in a pre-revenue company at a$20 billion valuation is presumably because there's trillions of dollars on the line.

1:26:17And so that's causing individuals to say, hey, I could go join this company. I could go on a vacation or I could raise a one on$5 billion pre-seed round. And that's pretty sweet. Let's go. And so I think there's actually somebody else that's joining Mira's company. I know Dylan Patel was tweeting about this saying that he was very excited about the startup and thought that the talent aggregation was very impressive. Still, I'm mostly excited to see like, how is the product going to take shape or be different? Obviously, there's a lot of difference. There's already fragmentation at the model layer.

1:26:55Obviously, we see that within OpenAI's model dropdowns. There's reasoning models. There's web searching models. There's AI models, code models. And so - Using OpenAI right now is like opening up a drawer in your desk that you just use to just jam, you know, jam a bunch of old papers in and you're trying to look for one and you're like, just sometimes you want to just give up and, you know, slam the door back shut. But that's easier to figure out how to name your products than. I was thinking about how messy the naming scheme is. And I was like, this is bad. But at the same time, I think it could be worse, honestly, if they tried to give them all like cutesy branded names.

1:27:34Like imagine if it was just like apple, pear, banana, like, like sprinkles, like ice cream. And you'd be like, which one's which? At least I know like 0103, like 03 is a bigger number, therefore it's better. And I can at least like click on that one. And I'm sure OpenAI's plan internally is you won't pick the model yourself. You'll ask what you'll tell it to help you with something and it will just route it to the best model based on the ask. And so, yeah, maybe it's like, you know, very temporary. Yeah, it seems very solvable. That doesn't seem like an intractable problem whatsoever. Anyway. All right.

1:28:09we got some questions in the chat. I just want to go through them. What you got for me? Uh, Elliot asked, can you guys go over Caterpillar earnings? Uh, that's a good question. I know nothing about Caterpillar's business, but I love, uh, I love, uh, I love the products that they make every time I drive by one with my son, he, he, he calls it out. So maybe we should do Caterpillar earnings. Yeah. Yeah. I love that idea. Ben, can you drop that in the group chat for, uh, for deep dives? We'll be actually, I would love to figure out how Caterpillar's benefiting from all the AI CapEx. Cause you could imagine if you have, you know, hundreds of billions of new data centers that need to get built, like that's potentially, you know, a big growth channel.

1:28:50So great call out Elliot. We'll look and do it. And if it's relevant, uh, cover it. Amit says, when's the Anduril IPO? Uh, I don't think we can, they said not this year publicly. I think they said not 2025. But I'm sure there'll be plenty of exciting news between now and then. Stay tuned. And we promoted Anderil job posting yesterday. So, you know, if you want exposure pre-IPO, you got to put in the work. You got to just go to work there. The underrated way to get allocation into these companies is just go in and exercise your options immediately. Yeah, yeah. No, it is probably easier right now to get a job at Anderil than buy Anderil secondary.

1:29:34I was scrolling their jobs page. They have like a thousand jobs up. It's crazy how many jobs they had. I thought it would be like, oh, like a quick scroll. It was like pages and pages and pages of jobs. So get on your grind, polish up that LinkedIn, polish up that resume, start, you know, being a reply guy to every employee. And in your cover letter, tell them the technology brothers sent you. uh another question from fsfsffrf which i presumably was he just slammed it in there to try to get a username going he says is this live and the answer is yes we're speaking to you live now there's maybe a five second delay but very live uh we got another question from 5523 what are your thoughts on the uk's prospects for the future tech wise um that's a better question for Harry Stebbings.

1:30:23He's on a mission to make Europe great again. I saw him wearing a shirt that looked, that had that on it. And yeah, I can't really like speak too clearly. I mean, I think that Europe has a lot of issues around tech regulation right now. I think part of the reason that the Figma deal didn't go through was they were facing pressure from European regulators, even though Adobe and Figma are both US companies. They seem to have a number of issues. Looking at the data, there's maybe only a small handful of companies that have been started in Europe in the last 50 years that are in the top 100 biggest companies in Europe.

1:31:06And so that just tells you that it's a tough environment to innovate in and create companies from zero to one. And so there's been some successes. the standout successes are the neobanks. I forget the name of the one that has done well over there. But overall, you know, and the other thing is Andreessen actually had made a big push into London around crypto because London was historically a finance, a global sort of financial center still is. And they the UK government was making a push to create more friendly regulations around crypto and Andreessen opened an office over there for that. But overall, I would say that if you want to build a high growth tech company and have the best chances of success, you're still better off, you know, moving to America or the Bay.

1:31:52Yeah. If you look at the, if you look at some historical data, I think from the top 20 largest companies by market cap, 100 % of them have been created in U S and China. Yeah. I think it's like, I think it's like 80, 20 U S. Um, and so, uh, unfortunately Europe's just, it's just not in the conversation. Um, you can certainly do entrepreneurship there and start companies. But in terms of these generational world changing power law companies, they just aren't coming from anywhere but the US and China. Yeah. Even China is making it difficult by in many investment structures, the founders have to like, basically put their house on the line where if the company doesn't work out, you know, the investors can can claw back.

1:32:38Yeah. And I mean, you can actually just go to Silicon Valley and start a company, raise money and grow and become an American citizen. This is a thing you can do. You just can't do that in China. Like there's never been a Chinese company that was not founded by a Chinese national. Yeah. Just impossible. Last question for now. And then I got a question for you, John. Also from FSFS. Is Technology Brothers a VC firm? Great question. We are not. We are the number one daily show in tech. John and I do invest. Though I've invested in 50 plus private companies. They're on my website, plus a number that haven't been listed yet.

1:33:15And John's done a lot. And John also is an employee of Founders Fund, which is a small boutique venture capital firm based in here in the United States. So last question. This is for you, John. What's your sleep score? Oh, yeah. Heard you had a rough night, but I got to put you in the spotlight. I don't have my eight sleep right now. And it's a real tragedy. gotta start traveling with it just take it on the road but what do you got so thursday was fantastic i was at 96 100 quality little low on the on the routine but last night was rough uh 90 quality for a final sleep score of 84 it's good but it could be better i'm trying to get to 100 every single day it's it's good it's probably uh probably better than the doge team which probably put up like three hours last night and uh you know on the pods but but probably still not putting up big numbers on the sleep score side but the experience has been fantastic i've been super enjoying my wife's been loving it it's just been amazing overall and i can't say enough good things about it i've really been loving it awesome all right well let's back to the timeline get back into the timeline we had a post from will monitis friend of the show underrated how much of the success of early stage companies are determined by vibes and not fundamentals.

1:34:40Better vibes mean people will give you cheaper capital, work harder for less money, give your product the benefit of the doubt. Once the vibes go negative, I'm not sure we actually have the full quote. It says, you are done. And then Lulu quotes it. This is timely because, so yesterday, Avi from Friend hosted a space because one of his competitors had a co-founder leave. And usually when a co-founder leaves early, people don't tend to jump off rocket ships. And so a lot of people were looking at that overall and saying, you know, hey, there's probably problems internally or maybe the product's not doing as well as they hoped.

1:35:28And so Avi knows how to get attention. He immediately jumped on and hosted basically a space to kind of I didn't listen for like two seconds, but imagine it was a little bit of dancing on on their grave. But yeah, I think Lulu's been like this is a part of her whole strategy with her firm is. And so she's adding to this. She said people think the purpose of comms is to release news and it is, but primarily it's for setting the vibes. And so I think that's very well said. It's really difficult to do. A lot of people associate vibes with taste because you have to have good taste to be able to consistently put out ideas and visuals and product releases and know how to leverage news and sort of make the story about yourself in some way.

1:36:18And so it's very difficult to do, but it does give you a massive advantage. You know, I experienced this with my last company. I would have people DM me that were incredibly talented. And like Dylan is a good example who led our marketing and partnerships effort. He was at a bunch of great companies. He was very early at HQ Trivia, helped blow them up, then was at Postmates. And he reached out to me and said, hey, I want to talk. I haven't seen vibes like this since early HQ trivia days. And then I pretty much immediately gave him an offer to join. And he's still a super close friend and went on to build a company called Crypto the Game and sold it.

1:37:02And so somebody like that, he explicitly came because he was just like, the vibes are good. I want to be a part of this. And so both Lulu and Will are very correct here. Yeah. And also, I mean, the important thing here is the early stage companies. In the late stage companies, once you've established the monopoly in the business and you're kind of Lindy, the vibes can go out the window and the profits can still print. Yeah, yeah. But, you know, we just talked about Amazon earnings. I would argue that Amazon doesn't have great vibes right now. The sellers hate Amazon. Users aren't excited about it.

1:37:36you know like the stock's up 40 is what i'm saying like yeah but i'm saying i agree with you but it just it just matters a lot less like it matters less yeah it matters less it matters it matters so much in the early stage that's yeah like sales salesforce has terrible vibes for our corner of x but people like to go to their conferences and like listen to uh pop music you know uh let's go to grant he says as a rule of thumb you should spend more time writing the prompt than it takes the LLM slash agent to do it. I think that's a really good tip. And I think I got so burnt out on the here are 10 sick hack and amazing prompts and amazing tricks to get something out of mid journey or open AI or GPT three that I kind of got bored of all that and stopped paying attention.

1:38:31But with deep research, I've noticed that getting the prompt right really, really matters. But it's not just the actual instantiation of the words, the idea, like what thesis are you bringing? What question are you trying to answer? What would require the agent actually to go and do a bunch of work and pull a bunch of data and then setting it up for success there as opposed to just, oh, if you use this word, you'll get a better result. Like for a long time, the prompting was like, think step by step or ignore previous instructions. And they were these hacks. And I think we're kind of post that era and now into an era of, okay, you actually need to think critically, bring an interesting thesis and the AI agent deep research or the reasoning model will help you substantiate that and support that with data.

1:39:17Yeah, I think I look at this and I draw a connection to when you're working with junior level employees, it ends up paying dividends if you spend 10 minutes or a thoughtfully crafted email explaining exactly what you want, how you want it done, what success looks like and things like that. And so you're just going to get a lot of better results out of that. And so I think what Grant is saying here is a good rule of thumb. I also think that this will continue to change as the models understand you better and they just get faster, right? Yeah, totally. Because eventually it'll be like nearly instant where, um, you know, yeah, you're going to want to spend a minute prompt, you know, writing the proper prompt and then it'll be something that's like, you know, completely instant.

1:40:05So, um, I mean, Dwarak Ashpatel was talking about this too. He was, he said, he was saying that, you know, we don't spend, we will spend hours training a junior employee, but we refuse to spend 15 minutes really working on a fantastic LLM prompt. And then he shared a prompt that was really robust, all for Gemini to help revise his show transcripts. And it gives bulleted lists of what he wants, what he doesn't want. It even includes an example of before and after, exactly how to transform the data. And he really set the LLM up for success. And now he has a template that he can use for every single show, drop it in there and get a perfect transcript every single time.

1:40:43And that's just not gonna happen if you just go in there and say, hey, clean this up, because it doesn't really know what you want. And so, yeah, spend the time to write a great prompt and you'll get great results. Let's move on. This is some drama in the hard tech world. Haas, the West's largest CNC machine tool provider, was found and charged for using shell companies to sell machines, parts, and services to China and Russia. And Cody James says, surprised to not see much news covering this. and Chris power over at Hadrian gives it the throw up emoji. Hit them with a throw up emoji. I mean, so one, I saw this news and my first thought was, so this is why the Haas F1 team is absolute trash.

1:41:32It all comes back to karma, right? You know, they, they were trying to be worthy American team, you know, cheer, cheer for us, you know, support, you know, support American F1. and meanwhile you're selling cnc machines out the back door to our uh enemies and i don't appreciate that and so i will um you know the next time i'm watching a race when they're getting you know uh 19th and 20th place and maybe not even finishing a dnf i won't feel so bad maybe it's Maybe it's, uh, but Hey, going Cadillac, Cadillac F1 team. It's kind of, it's happening. I'd like to see, I'd like to see Chris power come in and, um, you know, LBO pass and, and, uh, you know, make, make has great again.

1:42:23Yep. It'd be great to see. Uh, yeah, we got to dig into this. I mean, it's pretty, pretty crazy to actually go and set up all these shell companies. uh usually when this happened it's kind of uh like we we were more or less bamboozled by a shell company that was set up by a foreign adversary we didn't do our homework this is them proactively trying to sneak stuff out the back door so bad uh so bad yeah and uh you know if if they keep doing stuff like this i wouldn't be surprised if 10 years from now somebody makes a Haas, you know, cryptocurrency.

1:43:02Yeah, rough. Who knows? I wonder how big the fine was. I wonder what the punishment will be. We'll have to dig into this more, maybe do a full history of Haas. That would be a very interesting deep dive. I mean, at one point, certainly a fantastic hard tech company, probably lots of lessons to learn. Just got to keep the moral compass at True North. Yeah. Let's move on. We have a promoted post from Jeff Huber over at Chroma. He's hiring a chief of staff to work directly with him. Highly recommend working with Jeff. Great guy. They make a, they're launching Chroma Cloud, and they're looking for chief of staff to help them build and scale the business.

1:43:41You'll work directly with Jeff, the CEO. This is Chroma's first dedicated go-to-market hire. Your scope and responsibilities will reflect this. This is an exceptionally demanding high-velocity role. He's looking for five years of experience in technical product, business operations or strategy roles, strong understanding of product-led growth principles, data-driven mindset, technical aptitude, excellent verbal and written communication, and a great taste for design. Bit of a corporate athlete role, really. Love to see it. Jeff's also a smart guy. Look at him, putting his link in the second post.

1:44:18Numbers, that's the kind of guy you want to work for. You don't want to be working for somebody that's putting hashtags and links in their primary post. Bad news. Not a lot of alpha there. But anyways, great opportunity and great company. Yeah. So Chrome is a vector database, very useful tool in terms of AI and AI applications that maybe people don't often think of. the really cool application that I was hearing about was the ability to just stuff a huge transcript or huge document into Chroma and then run vector similarity with other documents. So you can imagine like there was a founder that was using this in a dating app.

1:45:02They have a huge intake form where they ask you all sorts of questions about what you're looking for. And then they can actually search over all that text for someone across the network who might be a good match instead of just trying to match you based on height and you know whatever other you know hard-coded metrics you you code in they can actually look really understand the responses who you are as a person what you're interested in your writing style and then match you across that and that's something that can that can happen in a uh in a rag uh retrieval augmented generation context very easily and so uh chrome has been on a tear they raised a bunch of money and jeff's an absolute dog.

1:45:41So I recommend hitting him up if you're interested. Let's go into some debate about Deep Seek. I love this post. Bojan asks, do people still use Deep Seek? And Nikita Beer trimes in with some evidence. He says, yeah, my cousin Jimbo Beer in West Virginia can really see the quality improvement over GPT. He heard about DeepSeek's performance from the town sheriff during Appalachia AI happy hour. And I love it. And Prashant there says, can confirm I host the Appalachian AI happy hour. And of course, Nikita is trolling. John, you normally leave me to botch words, Appalachia. oh yeah Appalachia uh uh but uh but yeah this is what we were saying the whole time the chart was clearly not not fully real you know I'm sure there was real I'm sure there was people coming from TikTok but uh I don't uh I haven't heard anybody outside of of teapot using deep seek, excited about it.

1:46:54I haven't seen on Instagram, haven't seen a single post about it or any type of content at all, really. Yeah. And Nikita broke it down really well. TikTok's been warming the trend. They're doing a ton of promotion for this. They're trying to get more people to install the app, but they don't really have any viral loops yet. And it doesn't make sense that there would be a massive uptake that would be durable or sustainable when you're going to be able to get similar results in WhatsApp and Instagram. Like you go to the Instagram search bar and you're talking to to a pretty solid LLM in the form of Llama and it's going to get better.

1:47:32The other thing that's interesting is I don't think the average chat GPT user cares about reasoning or would even really know how to really put it to use. So for them, if you say, well, I have chat GPT on my phone, I'm on the free plan. I get some value out of it. Oh, there's also a Chinese app that has Chinese characters as the developer name in the app store that I can't even read. Oh yeah. I'll just stick with chat GPT, please. Uh, it's, it's not a really good sell at the moment. Yeah. And as much as we make fun of the chat GPT app as being confusing with all the models, like it is a more robust and better app than deep seek just in terms of functionality sort of image generation baked in and charts and all these different PDF uploads we joke about.

1:48:20But it really is a more full-featured app just because they've been developing it for several more years. And so DeepSeek is kind of like a rough clone. And there's a free version of the ChatGPT app that it's going to be hard to get people to uninstall that. And then there's also Google AI summaries. And AI is popping up in all these different places where there's already distribution. So not super bullish on Deep Seek sticking around and breaking through in the way TikTok did. Yeah. Especially with all the pressure. Here's an interesting post from Cody James. He says, we should bring back How It's Made.

1:48:55This show did more for America's skilled workforce than anything else. Post clearly resonated. 1.5k likes. And did you watch How It's Made ever? I'm trying to remember. It sounds really familiar, but it might be slightly before my time as an almost Gen Zer. It was on, I think, Discovery or one of those kind of the Mythbusters network. And they would just show you, OK, we went to a factory that makes bubble gum. And we're going to show you the ingredients going in, goes into this machine, gets extruded. uh really nice cameras really uh like lots of slow motion just uh really nice voiceover and some really cool music and it was just really relaxing really fun to watch i loved watching this and it actually came in in in super handy when we were building the first uh uh when we were uh planning to build the first uh manufacturing line for lucy gum nicotine gum i literally watched how it's made on how to make gum.

1:49:59And I was like, okay, taking notes, like, okay, it goes in this machine, goes in this machine. And it's remarkably like making pasta. You like you, you have a bunch of ingredients, you need it into a dough, then you extrude it, then you need to cool it and kind of bake it down. And then you coat it. And so, and it's just like, it's really just like four or five machines. Each machine has a couple buttons and yeah, you can hook it up to a computer, but really it's just like, if you know how to turn it on, you know, the right settings, You know that, oh, if you put this flavoring in, you're going to run it for a little bit longer or you're going to need something else in there.

1:50:33And this is how pretty much every product in the world is made. Yeah. And I completely agree. We're getting a light version of this with the Jason Carman documentaries. I got a chance to do a version of how it's made with George Hotz because he has brought his entire manufacturing workflow in-house for Kama AI, the aftermarket. That's crazy. self-driving kit. And so he has a machine that prints circuit boards and then a machine that puts transistors on the boards and assembles them. And then he takes it over to another place. We need George to get into the drone market. For sure. That would be amazing.

1:51:09I, yeah, I mean, these, there's so many companies that need to be built. And, and unfortunately it feels like far too few entrepreneurs that have the will and the ability to actually go and execute at scale and, and really get addicted to the grind of, okay, we're going to sell more and more and more and grow, grow, grow. Uh, we, we keep coming back to like, Oh, there's a problem. Like, can Elon build a company around this? If not, we're screwed like Intel. Oh, it's gotta be an Elon project or no one. And, uh, as much as I love Elon, it's like, it's kind of sad that we don't have like a deep roster of Elons that can go and transform companies.

1:51:46Um, but hopefully the next generation is it. Hopefully they get inspired by how it's my B2 listeners, uh, build different ironically are. Yeah, that's true. Yeah. Um, awesome. Let's move on. Uh, this is fantastic news. Palantir has officially reached mega, mega cap status today, surpassing a$250 billion market cap. The stock is absolutely ripping. It's massive. And Gary Tan quote tweets it and says, I designed the logo for a mega cap 19 years ago. Overnight success sometimes takes two decades. That is so crazy. Insane. GT, GT, GT, the absolute, absolute dog. And Ty Morse down there says time to update this thumbnail.

1:52:37it's a vlog of Gary saying that he, I believe he didn't stay at Palantir long enough to get all of his shares. And he says, it's the decision that cost him 200 million. Pretty sure Palantir was worth like 10 billion back then. And so you might need to add another zero to that. But, you know, Gary's obviously gone on to do incredible things and, you know - Make it all back. Make it all back. Make it all back in the capital markets game. And it's still just like a fantastic story. and a really inspiring story for early employees and a tale about joining a rocket ship, identifying the right company, getting in and just riding it all the way to the top.

1:53:19And so Gary's been super, super generous sharing his insights on his YouTube channel. Now he's over at YC sharing more of those. I'm sure he feels like he hasn't told these stories enough, but I'm glad that he's resurfacing some of these stories because they are fantastic. Insane. uh oh this is some amazing breaking news uh Saudi Arabia has asked the Oscar winning composer Hans Zimmer to work on a new version of its national anthem I am so excited about this this is this is just incredible it only happens in Saudi Arabia uh and I love that they're just like yeah we need a v2 like no we should be so this this we should all we should commission new versions of the national anthem you don't actually have to be the actual country to commission a new version of your national anthem we could do we could do a you know i love it a doubt to make a new uh you know national anthem right let's get those crypto guys to uh to put their you know dollars to work and create you know a hard asset like a like a banger version of the national anthem but i love it i love that hans is is uh partnering up with uh the kingdom uh it it's gonna i cannot wait to hear this track we'll do it live um we'll we'll play it live on the show react to it um maybe you know we're gonna be in the middle east later this month maybe we can pop over and do a little listening party yeah um but uh it's great to see yeah i mean it makes sense like like there are plenty of national anthems that are iconic uh you know people obviously know the american national anthem but most people also know oh canada and uh and there's a few others that stick out in people's mind.

1:55:02Saudi Arabia hasn't charted top, hasn't top charts in a while. Most people can't sing it from heart. And so get Hans Zimmer in there. He did inception. He did the dark night. He did, uh, Oppenheimer, I think maybe that was somebody else. Um, black Hawk down too. Oh, he did. There's some crazy tracks on that. Uh, yeah. Yeah. I wanted to sound like inception really dramatic saudi arabia welcome to saudi arabia hans zimmer i mean i hope they play this at the airport just on repeat what do you think the actual cost to get an okay track from hans where he kind of looks in the back catalog he takes out some stuff that never quite got published yep because it'd be kind of cool to get the technology temple anthem you know a track that we could play while we're prepping the show and the studio yeah i think it's probably seven figs it's probably up there but it's doable for the right price um hans were also i mean people uh uh or john williams has gotten a lot of criticism for kind of like remixing gustav holtz into star wars and so you know a lot of it's just what what uh you know you're standing on the shoulder of giants what what building blocks are you using to create the next classical banger um but hans Zimmer's great he I think he did Dunkirk and uh he's done a lot of interesting stuff with like burying the theme of the movie within the song and so yeah I I'm genuinely very interested to see how he brings out the uh you want to the spirit of Saudi Arabia through in the track I'm excited one one thing that's interesting so in UFC when the the when the UFC goes to the UK yep fight on uh like basically like eastern standard time yep and so the fighters have to fight at like 5 6 a.m so they have to stay up the entire imagine pulling an all-nighter and then having to fight a title fight like it's the most brutal thing ever when they go to uh and there's a lot of english fans and so people have complained about that and said hey we have a lot of fans here you're making the fans and the fighters stay up the entire night to watch these fights in um when they fight in saudi arabia and the uae they fight perfectly at prime time and uh locally saudi time wow so so there was a ufc card i think it was last saturday yeah it was saturday and uh started at 9 a.m pacific it was actually kind of nice for me that's great it's like f1 f1 you're always up early and if you're in california but the saudis absolutely mog the the british fans by being like no when you have enough leverage they'll do it at whatever time you want it's a bad day to be this uh to be a sports fan if you're not saudi aligned like didn't they take all the golfers to this to the south yeah that was i guess i guess the pga and and trump are in conversations now bring it back we'll see yeah uh well let's move on to a promoted post from Christie's Real Estate.

1:58:04There's a breathtaking country chateau near Brantome set on 134 acres of tranquil parkland and forest. This would be a great pickup for anyone in the audience who's looking for a chateau. I really liked it because it has a lot of versatility. You can see massive turrets, multiple stories, looks like a castle, and it has a pool. great turrets on this one i mean not enough people take into account how many turrets they have when they're shopping for a home and it's just such a amazing feature one concern i have now even talking about this is that our friends over at wander kyle and john uh they listen to the show and i'm guessing that that they're listening live and i'm guessing that they're bidding oh yeah already and so uh uh if you're not listening to the show live good luck uh picking this thing up because the wander guys are on a mission to get the the best vacation rentals across the entire world in every single market so yeah so let's move through to the next slide uh we got some close-ups of the turrets we got the pool and you go inside and it's just loud opulence everywhere you look amazing it looks like it could use a little bit of this fireplace a little bit of a maybe not a remodel but a sort of refurbishment i think you throw an eight sleep on the floor and you're good.

1:59:24What else do you need? Yeah. All you need is an eight sleep and a dream in your Renaissance castle. A great place for offsites. Great place to have the whole crew while you're building the startup. Get a Chateau, the country Chateau. So yeah. If you're looking for a Chateau, go to Christie's real estate, tell them the technology brother sent you. And then, you know, while you're not there, throw it on wander, make it a little couple extra bucks. there you go that's that's the real play yeah uh all right i threw this in okay i'm tearing up right now okay break it oh yeah this is rough so um anyways so the news is that uh every quicksilver billabong and volcom store in the united states will close and uh it's very sad uh we had a we of Quicksilver in Malibu that shut down.

2:00:18And if you're a surf apparel store and you can't stay in business on PCH in Malibu, that just says that there's so little demand for surf apparel now. And so to give some backstory, because I don't think I've talked about it on the show, I grew up an avid surfer in California, went to UCSB because the surf was good to be on the surf team. And so I competed on the team and and I actually worked at a surf store in high school. And even then I could tell that surf culture was fading quickly from relevancy. It used to be that, you know, young kids in Ohio wanted to wear surf apparel because like surfing was cool.

2:01:00And then skateboarding really rug pulled surfing that became more of the sort of dominant culture, at least in apparel. and then streetwear was really the death knell for uh surf the surf apparel industry because kids now want to wear vape and uh you know supreme and you know nike and things like that and so anyways the surf the surf apparel industry is in a rough spot i mean i'm sure the brands will will survive it's just more so uh athleisure is the other big trend right so dads ads. dads might have been wearing Quicksilver shorts and shirts and now they're wearing Blue Lemon or um what's the other one that um i don't wear athleisure so i can't even think of it but um worry worry worry um worry is the other one yeah those those those brands like just really dominate that i mean there's a beautiful hawaiian shirt right there in that image palmer lucky should pick this up it can be his next mod retro yeah he should pick up quicksilver billabong volcom turn them around just you know start printing make it great yeah i'd love to see yeah and it's this i've talked about this on the show before my my um roommate in uh in in college pierce incredible surfer easily could have gone pro but his parents forced him to get an education and go to go to high school and college um but i i swear he he has he now uh is able he's in real estate He's able to work all over the world.

2:02:34So he's like constantly surfing in different places while having a real career. And so he's getting the last laugh as the sort of surf industry collapses. And a lot of the, you know, salaries that were paid to surfers historically for endorsement deals have dried up. And, um, so anyways, sad, sad moment, but I think a lot of people saw it coming out. Uh, art club says pouring one out for my seventh grade wardrobe. and josh steinman says jinko uh let's go to katherine boyle she says it's crazy how good the internet is right now best season ever and carter says x is the movie theater couldn't agree more having tons of fun on the timeline uh never a dull moment and there's ever a lack of content for our show yeah i i swear we've had consistently a new current thing every 48 hours this year and And it's fantastic.

2:03:28I mean, part of that is the Trump effect. In 2016, I kind of, I was in college. I had like work in my business. I just kind of like tuned out a lot of that stuff. But this is exactly what happened back then where every single day he just sort of dominates mindshare in the news cycle. And it's been interesting now having JD there and Doge and some of these like bigger initiatives that, um, and having that, that you have, um, the AI, uh, activity as well. It's just kind of creating this sort of vortex, this attention vortex. And, uh, yeah, I think, I think many people are saying it's just going to keep accelerating.

2:04:08I completely agree. I completely agree. Let's go to Brandon Jacoby over at X Jacoby. He's on the timeline. line he says i can't say why exactly but i need you to like this post and it's a banger he got almost 5 000 likes jacoby you know the steel man here is that oh he's a designer at x he's testing the algorithm he's testing some feature in prod no i think he's just farming he's just farming likes trying to grow his account trying to you know get that uh greater monetization check going print some elon bucks yeah this is this is uh this looks like insider trading on the on the creator payouts yeah i'm gonna use my i'm totally joking it's funny i asked i asked tyler and jacoby uh if if i could record live from the x studios or the x office in new york on uh on tuesday they've yet to get back to me they said they um they said that it's pretty tight security wise over there.

2:05:10So I can't have, I can't have newscasters just stroll in the halls recording, you know, live shows, but we'll see, maybe we make it happen. Yep. Well, whatever he did, it worked. And if you're ever in need of a pickup and a banger, just, just copy paste this, throw it in the queue, your notifications will be popping off. And, uh, you know, I think this is a tried and true format, almost like, uh, he was so real for this with a screenshot of a banger works every time. Uh, let's go to link link says day two of my one K to 50 K challenge. And the portfolio is down 99.75%. And link, which started with one K has$55 and 98 cents.

2:05:58And so, you know, you just need a, you just need a, a thousand bagger. Now, Originally, you needed a 50 bagger. Now you need a thousand bagger. You know, dust yourself off. Get back on the horse. I believe in you, Link. I want to see what happens. I know whatever's going on. I don't even know this app, but I know it's degen just from looking at the UI design. I know you're trading some sort of coins. That's Phantom. Okay. Yeah, there you go. And you know that this happened very quickly. And this is his 10th account, to be clear. Oh, yeah. You can see he's got a bunch of, bunch of wallets gone.

2:06:33So, I mean, yeah, most people would say, Oh, I'm down 99.75%. Uh, you know, it's hopeless. I should just give up now, but 99 % of people actually quit right now. And so take your 55 bucks, write it back up, hit the hit the timeline and run it, run it back up to 50 K. I posted, I hadn't actually seen this until you shared it, but, um, I posted today, uh, we need a McDonald's challenge in crypto where a bunch of contestants, they each get 25 K they've got some amount of time to turn it into a million. And if they're successful you know, great. If they, if they lose the money, they have to go get a job at McDonald's and just use those paychecks to pay back the 25 K, which I imagine would, would take, unless you got a nice, a cushy management role, it'd probably take you a year just putting the fries in the bag, bro.

2:07:27Yeah, I like that. That's a good one. Hopefully that pops off. Well, let's go to another promoted post from NASDAQ. Good friend of the show. Good friend of Ramp. Run that wonderful billboard that partnered with AdQuick. AdQuick talks to NASDAQ. They own that beautiful billboard in Times Square. They'll throw your brand up there. So call AdQuick if you're looking for some promotion out in the real world. It's the way to make an impact online. You go, you throw your brand on the big billboard. You take a selfie with it. Elon retweets it. Next thing you know, you got 10K likes. What you got, Jordy?

2:08:01I will, I'll put a little note out there. If you want to run an ad quick billboard in Times Square at any point in the next year, DM me. I'll spend 30 minutes with you on Zoom and work on the creative. I love doing out of home ads. And so a little free marketing support for you. And it's completely power law driven. it. Like the difference between just a few, what a few words on a billboard say completely change it from being, Oh, there's a billboard there too. I need to take a picture of that, put that on social media and it goes viral. And so you really can't under under appreciate and under count how important dialing in the copy and dialing in the messaging and the imagery and the design of that billboard.

2:08:45Because if your message is iconic and it resonates, it's going to go so far and you're going to get incredible ROI on that out of home spend. And so highly recommend it. But the reason we're talking about this is because NASDAQ, they went to the World Economic Forum in 2025 in Davos. We've said that we're going to make Davos great again as soon as they give us a call and let us host it. Executives discuss the economic outlook and cutting through complexity with media and industry leaders across BCG, Tenio, IIF, and CNBC with special guests, Simone Biles and Man United. You're only going to get that type of lineup at Davos and Nasdaq is there to host it.

2:09:28Low Tam banger, at least at the time of screenshot, four likes, few retweets. I guess they're making it difficult to engage only accounts that Nasdaq follows or mentioned can reply, but I know why that is. Yeah. Coins. You know that, you know that anytime they post someone was like, let me put a pumped out fund on here. If you want exposure to NASDAQ, W-E-F, get this token. Not financial advice. Do not buy anything ever. But we love NASDAQ. We love companies on the NASDAQ and we support the NASDAQ. So thank you to NASDAQ. Let's move on to another Ramp post. This one just from a fan. They say, I would love to see Ramp produce a Spotify-esque wrapped report.

2:10:21Some ideas. Your top reimbursers for the year. Simon spent X dollars on his Ramp card. The engineers expensed X dollars on Postmates. I think this is a great idea, not just to go viral, but actually to understand how the money's flowing around your business. A lot of times when you close the books, what you're doing, digging into below different cost line items on the P &L. It's to find weird things that are happening. And a wrapped type product would be a great user experience to make it actually fun to dig into the different parts of your business and understand, oh, like what's really driving revenue growth?

2:10:59Where's our marketing spend going? Where are we spending too much on travel and entertainment or office supplies. Any of that stuff could be really well surfaced through AI, through even just normal algorithms, just figuring out what's going on, showing me a couple trends, but just surface this, you know, ramp is, you know, I like to think of it as machine God in your CFOs Chrome tab. And we need more products that make it easy to understand what's going on in your business. And that's why we love ramp here. ramp ramp onto the next one ready to do that one uh should we talk about sahil laviniga lavinga lavinga let's do it he says no longer hiring junior or even mid-level software engineers and he breaks down how many tokens are in each of his code bases gumroad is just two million tokens Flexile is 800k, Helper's 500k, Ify's 200k, and Shortest is 100k.

2:11:59Both Claude 3.5 Sonnet and 03 Mini now have context windows of 200k tokens, meaning they can now write 100 % of our Ify and Shortest code if prompted well. It won't be long until AI will be writing all the code for Helper, Fixile, and Gumroad. My guess is by the end of 2026 when these models get bigger and their context windows get bigger. And so his new process, step one, he sits and chats about what he needs to build, doing research with deep research as he goes. Then he has AI record everything and turn it into a spec. Then he cleans up the spec, adding any design requirements or other nuances.

2:12:32Then he has Devin coded up. We're big, good friends with cognition. Love that Devin's helping out here. And then step five is QA merge and auto deploy to prod. And so he is just having a great time with AI tools. And it's great to see a CEO of a company that has a couple of different products, a couple of different code bases actually break down the workflow. One cool thing here, you talked earlier about how a CEO like Zuckerberg just says, like, this is my company. I have control. I'm just going to do everything that I want to do under this umbrella. Yep. Sahil is doing the exact same thing with Gumroad.

2:13:09He's basically built out a number of different products or business lines under Gumroad, which is, um, yeah, I'm, I'm sure everybody's familiar at this point, but an easy way to sort of buy and sell digital products online. Um, and so he's now building out these other products that help him build gum road and then making those individual products available to everybody else. So they have, um, some type of like, uh, contractor payments product. They have a help desk product that he's talking about here. So I'm pretty sure he raised VC money very early. He started the company very young. He was, I think, an all-star intern at Pinterest and then raised a bunch of money.

2:13:47And then I think he was able to buy out the VCs at one point and kind of get more control over the company or something. And now he runs it kind of like a lifestyle business, but that's not meant to be a pejorative. He just has control and he's been building exactly what he wants and growing the business in very interesting ways. Unfortunately, documenting it all right in public. He's not quite building it, but there's a lot of good stuff that comes out of his account. yeah and he he did a very high profile crowdfund i think in must have been 2022 that's right but very small very very small shareholder in that oh cool so anyways excited to see where he takes it well let's move on to trey stevens uh answering a very interesting question because trey worked at palantir with alex carp and chase asks trey what was it like working day today with carp i'm so curious and trey says well one time uh my buddy and i were who uh who went to notre dame and georgetown respectively were with alex carp after a great customer meeting and during a moment of silence he said great job guys you think we should be recruiting more people like you from second tier universities that's mostly what it was like i think that's so funny uh carp just talking trash, talking trash in the boardroom, but it worked out.

2:15:05He got great results from his crew. Yeah. And he's still, still talking smack this time from earnings calls. We love to, we love to watch. Yeah. And, uh, Tim Ronan, who was there says Carpe Diem with a K and Trey says, you remember this, right? I don't think I was hallucinating, uh, such a funny post and just a great culture, you know, just guys, guys being guys talking trash in the office. I'm sure it makes everyone grind harder, way better than, you know, oh, you know, my boss can't make fun of me. I couldn't possibly withstand that. No, you know, you got to have some sharp elbows. You got to have some fun sometimes.

2:15:44So we'd love to see it. Let's go to Phil Mickelson. Phil Mickelson said, I would, is this the actual Phil Mickelson? I think so. He says, I would just like to say Sesame Street is an excellent show. Iraqis deserve to have it. And although production costs were under 500 K shooting on site required$19.5 million to ensure. Okay. This is from the USAID stuff, right? Yeah. There was a USAID payment for$20 million to produce a Sesame Street for Iraq. Yep. And so Phil's just backing them up. He says Sesame Street's an excellent show and Iraqis deserve to have it. So I think standing up against - I talked about this on the Pirate Wire show yesterday and it's a pretty hilarious story.

2:16:35What's weird to me is like, why did they need to shoot Sesame Street in Iraq? Like, why not just take our Sesame Street, translate it and then just broadcast it over the air or just send them the files and be like, you can distribute this now. It's in your language. Instead, they had to localize it and shoot it there on site, which is crazy to me. I like how he follows this up and says, cause sarcasm is just completely lost on X. If you ever post something sarcastic, you'll understand, but Bill replies, I'm curious how many people will think I'm serious. And then he replies again, turns out quite a few wild story, but you love to see Phil Mickelson on the timeline, dipping back in, uh, not, Not your typical teapot poster, but still got it.

2:17:20Got some good humor. Still got it. Having fun. Wading into it. Let's go to Cointelegraph. Just in. Pump.fun hit with a cease and desist from Berwick Law ordered to remove 200 plus IP infringing meme coins. And, you know, I don't know. We could be a plaintiff in this case. Someone made a Pump.fun meme coin all about us trying to steal our idea and our branding. Well, yeah, to be clear, they made a coin. We had talked about LVMH being a good proxy for TV because we think a lot of people in tech are going to be big LVMH buyers in the future. And it was said purely for entertainment, but somebody turned it into a meme coin.

2:18:01And then in the description on Pumped Up Fun was saying that it was affiliated with us, which it clearly was not. But the funny thing about this post is that you can't just remove tokens from the blockchain. I guess PumpFun could stop showing tokens that have IP infringement. Yeah, they can block them at the Web2 layer, but not at the Web3 layer. Yeah, the tokens are still going to be there, still able to be traded. And yeah, it's just one of those things that companies have to... I mean, the NFT marketplaces went through the same thing where there were, you know, really crazy IP infringing NFTs and some were very offensive.

2:18:42And there were a lot of NSFW NFTs and a lot of spam NFTs. And so OpenSea and the other marketplaces created essentially views and filters on top of people's wallets to only show certain products. And then obviously only highlight certain products in the store. It seems like that's something that Hump Fund should do in the sense that there was a TikTok token and it had nothing to do with TikTok. It was just, and you had a guy from Trump's team or some organization funded, you know, I don't know, part of that whole world. He was promoting it. So I think that there is a real risk for consumers to, you know, think that it's somehow associated with TikTok and maybe they're buying it.

2:19:31But that being said, people need to do their own research and think, hey, this is a Chinese company that's under immense scrutiny right now in the US. launching a crypto token, even though they make billions of dollars a year? Probably not. Well, thank you to Berwick for your service. And we're happy to support you if you need extra evidence of what's going on in the IP infringing meme token space. We will be on the witness stand. Yeah, we'll call up Bernard Arnault. Yeah, get him in there. We'll come at it together. Yeah, no, I would be happy to take the stand next to Bernard in defense of valuable intellectual property.

2:20:11Let's go back to GT, the man, the myth, the legend, the president of Y Combinator, Gary Tan. He says, that is merely the average at YC these days, thanks to AI. He's quote tweeting, Yuri, who says, overheard from a Series C founder, we are experiencing YC level growth numbers 10 % per week. Let's go. I love it. Insane. Yeah, I posted earlier this week that YC is just such an obvious beneficiary of AI for a few reasons. One, you can just build things faster than ever. So the three-month period that YC typically takes place in, they can just make so, you know, these teams can make so much more progress.

2:20:53Two, teams need to hire less people because they can use AI to just be way more efficient. And then three, these companies are, so I'm going to give you four actually, but one is that the potential applicant pool has exploded so much because people that didn't think of themselves as developers can now develop software. And then the last being that now that it's cheaper to build software, we're going to want more of it. So I talked to a team yesterday that's doing something in sort of retail, more consumer packaged goods, and they're building software as part of their business because they now can do something with one engineer that would have taken them, you know, a pretty significant team historically.

2:21:38So very cool to see. That's great. And Gary, if you're listening, let the Technology Brothers host a live stream for Demo Day. We want to review all the startups, pump them all. Yeah, right now, they work with TechCrunch. Give us the press pass. Yeah, all we need is a press pass and we will put the companies on. Yeah, it'll be great. I think we make it very entertaining. We'll stream it on Axe, YouTube. We'll bring a lot of entertainment to YC Demo Day. Be great. Let's go to Zach. He says, if you're not coming close to peeing yourself at your desk, you are not obsessed with what you do. And this really resonated for me because oftentimes we'll finish recording and it's getting hairy, you know, it's getting close.

2:22:29No, I think this is just a sign that you're working on something that you're really excited about. If you're working on something that bores you and you, uh, and you have, you feel like you need to go to the bathroom, you'll just get up and do it if you're working on something that's stimulating yeah you know you probably let it go a little longer than you should so yeah and lucas here says dude's so locked in he got a catheter yeah zach said he'd consider it yeah dude is the definition of locked in yeah i i wouldn't be surprised if some of the doge boys got catheters in so they can just stay stay in front of the computer you know grind never stops the guy the grind never slops either perfect Freudian slip, I suppose.

2:23:11Uh, let's go to Josh Steinman, friend of the show been on many times, you know, him, you love him. He says this exactly. And the bridge from the idea to results, the gap is being cool and the solution is being cringe. And so you can't be afraid of being cringe to grind towards your results. Yeah. They're only going to be, you're only going to be remembered by what you accomplished. So if you need to be cringed to accomplish something, that to me says that cringe is cool. Cringe is bad. I mean, the flip on this, like cringe is like, it's such a charged word, but it's more just like putting yourself out there.

2:23:54Even just asking your first customer to go from a free trial to paid, it kind of triggers that cringe feeling of like, Ooh, I'm putting myself out there. I'm asking for something. I'm asking for a favor. I'm asking for something that I might see. I might, there's, there's like downside personal ego risk. And so I think that's really the abstraction layer here of, of even just telling your family, I'm going to start a company and I've never started a company before. That is in many ways, in some circles, like cringe, like, why don't you go be a lawyer? Why don't you go be a doctor? And, and you have to be willing to put yourself out there and take that, that ego risk.

2:24:33And that's what cringe is. It's not necessarily actually just like when people think cringe, they think like posting slop threads on X. That is cringe, but that's not what we're talking about here. I think you nailed it. You have to be willing to be successful. You have to take ego risk all day long for years, right? And even a very small example is us transitioning from a recorded show to a live show. It was a bit of a hurdle. Suddenly, everything we say is immediately public. Um, thankfully we never talk about politics or social issues, just purely about business and makes it a little bit, uh, easier, but, um, but yeah, you gotta be willing to be cringe and Simon even he's notorious for posting the same idea over and over and over.

2:25:21And a lot of people would say, well, it's cringe to say the same thing over and over, but he made it part of his brand. And I actually think it's cool now because he's just saying these ideas are important to me. I'm just going to keep posting through it. Yeah. He's gotten results. the man has gotten results it's fantastic uh we got a couple more posts in the timeline we got five more minutes uh lm over at kline perkins one of the holy trinity firms in the valley says product exec at startup in response to cogen tools i told my pms we are done with specs or prds prototypes only and so you're on notice pms if you're not building prototypes you're not going to be working for this product exec at this particular startup even though it's anonymized but interesting makes sense yeah i saw tj parker posts that the entire time they were building uh uh it was what was his company again pill pack yeah yeah pill pack sorry i thought i was getting confused the entire time they were doing that they didn't have any pms it was just designers and engineers um and so i think that there was at at some point over the last 10 years it was very it was almost like controversial to be like we're not going to have any product managers and I do think now um you know we still need people to um yeah it'll be interesting to see how it plays out maybe it's designers that code and then you have like operations people or vice versa but anyway the flip side of this is what do you do with your pms who can't build prototypes can't design can't engineer and my idea is you send them to the pool and you get of making TikToks, they go viral because it's insane rage bait.

2:26:57And then everyone learns about your company and they're like, wow, this company is so dysfunctional. You're playing 40 chess. Your PMs are now marketers for your company. Yeah. Your PMs just become your marketing budget. So however much you're spending on PMs, that's what you would have spent on ads. And if they bait people properly, uh, you'll get way more impressions than, you know, whatever you would. Yeah. And so PMs either you're in, you're in Devin, you're in cursor, you're in windsurf building prototypes or you're at the pool and you're opening up tiktok so there you go spms now and be now in between no in between stop writing prds prototypes only and that concludes the timeline for today we are on schedule two minutes early this is a fantastic show leave us five-star review on apple podcasts and spotify please leave an ad in your review we will read it out on the show and just thanks everyone who tuned in thanks for watching Thanks for being a part of this community, the brotherhood.

2:27:54The brotherhood. International technology brotherhood. And we will see you Monday. We have five more days of content planned next week. Lots of clips going out. Follow us on X, TBPN on X. Subscribe to the YouTube channel. Add it to your Spotify, Apple podcasts. Review us everywhere. Send it to your mother. We specifically have eliminated swearing from the show so that your mother won't be embarrassed to hear you listening to people with a potty mouth. I do have a rant about the free speech thing that's going on. There's a lot of people that are saying more and more aggressive things. And then there's a question about, will there be accountability for this?

2:28:37What's the free speech thing? And I just think going back, I need to coin something around this, Coogan's Law, but I've been calling it the, what would your mother think rule? And I think that in the age of truly free speech, You're not going to be, this is America. This isn't Europe. You're not going to be put in prison for your speech. And also the platforms are extremely forgiving. There's very little censorship now. You might get community noted, but you can kind of say whatever you want. And so it's incumbent upon you as an individual and as an upstanding member of society to not police yourself, but just act with honor.

2:29:11And the heuristic that I think serves us best is what would your mother think? What would your mother say? and so that applies to the language that you use, to the ideas that you espouse and just if you're being a nasty vitriolic person and your mother wouldn't approve of that, maybe you shouldn't do that at all. Maybe you should be going golden retriever mode and trying to be friendlier and so that's how I'll end the show. Well said. Thanks everyone for watching. Thank you. We'll see you Monday.