Mukund

One of the most substantive classes with @ChaseLochmiller at Stanford. We went deep on economics of the datacenter: - Where is the ~$650B of AI infra capex actually going this year? - Who's capturing the margin, who's getting squeezed? - How the bottleneck has moved from GPUs to power, and where it goes next - The economics of neoclouds

🚀1 million+ delivered bookings. And counting. 🚀 "InstaHelp’s trajectory is a testament to the trust our consumers place in the platform for their most immediate, everyday needs." - @abhirajbhal #InstaHelp

I'm seeing this pattern with CEO dashboards now. CEO A: - Spends $50K on fancy BI tools - Has dashboards nobody trusts - Argues with teammates about feelings instead of facts - Systems break every few months - Still firefighting in Slack and email CEO B: - Uses a Google Sheet - Tracks metrics weekly - Makes their team enter metrics manually - Uses green/yellow/red stoplights - Reviews the whole business in 15 - 30 minutes In 2026, guess who scales to $20M?

FULL INTERVIEW: @travisk joins TBPN to discuss his new company Atoms, physical AI, Uber, and more: 01:18 - Why he's been building in stealth for 8 years 04:32 - Atoms and the future of physical AI 08:10 - Creating a culture of builders 12:05 - Lessons from Uber 24:30 - The vision for physical AI and robotics 31:15 - Why humans will be the main beneficiaries of AI 38:20 - Mining, autonomous robots, automation 47:05 - Why Travis moved to Texas

Heard on TV Whether the Elephants fight or make love, the grass always gets crushed. India mkts is the grass

Jay Shah just nailed it. 👏🔥 x.com/middle451817/s…

The biggest flex of my life is backing this clutch monster all these years. Thanks for making us proud Sanju Samson ❤️❤️

Take a bow, Sanju Samson 🙇‍♂️ #TeamIndia | #T20WorldCup | #MenInBlue | #INDvWI | @IamSanjuSamson

Citadel elegantly murders the Citrini post here citadelsecurities.com/news-and-insig…

Takeaway from the Citrini backlash should not be that all marketplaces are immune to AI, just that DoorDash was a particularly bad example to choose. Defensibility is largely a product of how far they are to the right on this spectrum. The argument is "agents will just transact on your behalf and look for lowest price" This is massively flawed for two reasons in the case of DD: 1. Price is function of network density. You must be able to optimize routes and batch orders to win 2. Even if someone else could win on price, customers care about many other things (selection, quality, service, all of which DD has invested in heavily) So you can’t build a good agentic food delivery product without DD cooperation. And for obvious reasons, they will not cooperate. But it doesn’t follow that this will play out everywhere. The more heavily managed a marketplace is, the harder it is for someone else to cut in. There are basically 4 levels of marketplace: 1. Lead gen: just a list of suppliers 2. Transactional: also handle payment 3. Managed: also take on risk (returns, net terms) 4. Heavily managed: also manage service delivery Google has been trying to eat the marketplace profit pool for many years, and really only succeeded in taking most of it away from lead gen marketplaces. LLMs are another aggregation layer like Google, but with two big differences: search is much better, and critically, they can transact on your behalf. So LLMs should be able to push up one step farther in the stack, and take on transactional marketplaces directly. But is Anthropic going to try to do the final two jobs of managing risk or managing service delivery itself? Are they going to start accepting returns? Offering financing terms to buyers? Are they going to manage their own drivers or build their own logistics network. That seems very unlikely. As a result, managed marketplaces are largely safe. Marketplaces that do the hardest, most capital intensive, most scale-dependent stuff will get rewarded for it. I wrote an essay on this here: danhock.co/p/llms-vs-mark…

InstaHelp crosses 50,000+ daily bookings in under 1 year 🚀 From a small pilot in Mumbai in March’25 to reaching this milestone - the traction demonstrates strong consumer demand for reliable, on-demand housekeeping services. We are investing to build a large, high-frequency category that deepens platform engagement and strengthens long-term growth for @urbancompany_UC

𝘠𝘰𝘶𝘳 𝘥𝘦𝘢𝘵𝘩 𝘸𝘪𝘭𝘭 𝘤𝘰𝘮𝘦 𝘰𝘯 𝘢𝘯 𝘰𝘳𝘥𝘪𝘯𝘢𝘳𝘺 𝘥𝘢𝘺, 𝘪𝘯 𝘵𝘩𝘦 𝘮𝘪𝘥𝘥𝘭𝘦 𝘰𝘧 𝘶𝘯𝘧𝘪𝘯𝘪𝘴𝘩𝘦𝘥 𝘱𝘭𝘢𝘯𝘴, 𝘢𝘯𝘥 𝘵𝘩𝘦 𝘸𝘰𝘳𝘭𝘥 𝘸𝘪𝘭𝘭 𝘤𝘰𝘯𝘵𝘪𝘯𝘶𝘦 𝘸𝘪𝘵𝘩𝘰𝘶𝘵 𝘺𝘰𝘶.

I spent 100 hours over the past week researching, writing and editing the piece we just put out. It’s a scenario, not a prediction like most of our work. But it was rigorously constructed, dismissing it outright requires the kind of intellectual laziness that tends to get expensive. And we’ve released it for free. Hopefully you enjoy it. citriniresearch.com/p/2028gic

Born Cool. Built Retro. Eternal Swag. 😎 #KHxRK Tamil Promo ▶️ youtu.be/TROFVicWrTE @ikamalhaasan @rajinikanth @Nelsondilpkumar @anirudhofficial @DirRajivMenon @Nirmalcuts @KiranDrk #PallaviSingh @amritharam2 @valentino_suren @alagiakoothan @MShenbagamoort3 @arjundurai #RedGiant17 ✨

Name : Rajinikanth Age : 75 Duty : Aura Farming

Holy af, edits are already here 🗣️🔥 x.com/ittzmaddog/sta…

Hi Chellams, I don’t think I will ever recover from what has happened over the past few weeks. It all began with A. R. Rahman sharing one of my reels, a bit from Alex in Wonderland’s ARR segment on Instagram. That alone was more than I had ever wished for. That was enough. (1)

A 15-year-old girl immigrates to New Jersey from China. Doesn’t speak English. Her parents, both educated engineers back in Chengdu, are now working as cashiers and restaurant cooks. She gets a job washing dishes at a Chinese restaurant to help the family survive. She gets into Princeton on a full scholarship. Her reaction is so disbelieving she asks two different advisors to verify the acceptance letter is real. Then her mom gets sick, so the family opens a dry cleaning shop in Parsippany. Every weekend for seven years, Fei-Fei Li leaves Princeton’s physics department to run the register, handle inspections, talk to customers, manage billing. Monday through Friday: quantum mechanics problem sets. Saturday and Sunday: sorting other people’s laundry. She later called herself the “CEO” of the dry cleaning business. She kept running it remotely through half of her PhD at Caltech. In 2007, she proposed building an image dataset so massive her own mentor told her she’d taken the idea “way too far.” Pre-ImageNet, the entire AI field was working with datasets containing a few hundred images. She built one with 15 million. Most researchers at the time believed algorithms were the bottleneck. She bet on data when nobody else would. By 2012, a team ran a neural network on that dataset and halved the existing error rate overnight. AlexNet on ImageNet became the moment the deep learning era started. Every computer vision product shipping today traces its lineage back to that dataset. Fast forward to 2024. She starts World Labs. Four months in, $230 million raise, $1 billion valuation. Today, $1 billion more at roughly $5 billion. The bet investors are making: that the woman who gave AI its eyes with 2D image recognition is about to give it spatial awareness of the 3D physical world. Her new model, Marble, generates persistent 3D environments from text or images. Unlike video generators that fake depth frame by frame, Marble creates actual geometric space where objects stay where you left them. The investor list tells you everything. AMD and NVIDIA both wrote checks. When the two biggest competing chipmakers both fund the same startup, they’re telling you this workload is coming whether their competitor funds it or not. Autodesk put in $200 million and signed on as strategic advisor, which means they see spatial AI integrating directly into CAD and design workflows within 18 months. From dry cleaner to ImageNet to a $5 billion spatial intelligence company. Fei-Fei Li has now placed two bets that the rest of the field thought were too early and too big. The first one created modern computer vision. The second one is trying to give machines the ability to understand physics. If she’s right again, this is the last major unlock before embodied AI actually works.

Was listening to @winstonweinberg from Harvey talk about scaling on 20VC with @HarryStebbings and he said something that should make every founder pause. For the first ~2 years, when he did revenue projections, he wasn't doing the basic capacity math. How many AEs at what quota, with what ramp time, meaning he needed to have hired them by when. His words: "I'm dead serious… I never even thought about that." This is Harvey. One of the best execution machines in B2B AI. And even they had a period where the spreadsheet wasn't connected to reality. You don't "manifest" revenue targets. You staff them. Revenue is an output of inputs. How much pipeline are you creating? How many reps can sell it? How long does it take them to ramp? What's your win rate? How long is the sales cycle? Can your team actually implement what you close? And are customers staying and expanding? If you're missing one variable, the plan is cosplay. The best founders I work with treat RevOps like engineering. Instrument everything, build the model, run weekly retros, fix constraints and repeat. They build their company on top of a strong foundation with a focus on the math Day 1. One uncomfortable question every founder should sit with: "If we had to hit next year's number with no heroics, what would we need to hire and by what date for the math to actually work?" Do the math early. It's cheaper than learning it after it's too late.