“don’t train your own model” is common ai advice. it's wrong. your token bill's the proof. today, we’re excited to launch castform into open preview. castform is the easiest way for you to train your own model, on your own data. open-weights models are performant and much cheaper. when trained on your task & proprietary data, they beat closed models. the thing standing between you and that was weeks of plumbing & years of ml expertise. with castform, model training is as simple as prompt engineering. @castformai bring your agent traces or raw corpora. castform turns it into training data, picks the right algorithmic recipes, manages gpus, and gives you an ide to watch and chat with your model as it learns. see what you can build with castform👇

Peter Thiel on the type of company more startup founders should build Thiel first emphasizes his belief that when starting a company, you should always ask: “Can this company become a monopoly?” He then lists three of the most common types of monopolies: Super fast distribution on a very thin product (e.g. Twitter) A technological advantage that is continually built upon with iterative improvement and compounds over time (e.g. SaaS software) A truly brilliant breakthrough (e.g. Bitcoin) But he argues that there’s a different monopoly category that’s continually overlooked: “A different modality for innovation that we do very little of and we don’t even recognize as an important category is what I would describe as ‘Complex Coordination,’ where you take a lot of different pieces and the challenge is to coordinate them into something new.” Thiel continues: “This is the thing that’s maybe 180 degrees antithetical to the Lean Startup ethos. It’s complicated. You have to put all the pieces together in just the right way. I think this is on some level what really drove Apple as an innovative company in the last decade… What was new about the iPhone? There was no single component that was new. It was just that you put all of these things together in just the right way… and once you built it, it was actually super hard for people to replicate. You had an advantage for many years.” He points to Tesla and SpaceX as more recent examples. “There’s no component to the Tesla that’s actually that new. It’s just that you put all of the pieces together. You re-engineered the whole distributor network. It was this complex coordination that made it work. There’s like this lost art of accounting where you figure out how much things cost and add them all together. And Elon has discovered this lost art of accounting which no other people practice.”

RT @lightsprintai: We’re live! Try us out on lightsprint.ai 🚀🚀🚀

Lightsprint.ai has created hosted GStack, so it's the fastest way to get going on creating new things

Our planning engine is now live! We've been pretty blown away by how it's able to let agents one-shot an entire feature 🤯. Follow, like and repost and DM us and we will send you free credits to try it out!

Our non-technical intern shipped new features to LightSprint yesterday. She described what she wanted in plain English. LightSprint made the change on our live codebase, gave her a preview URL to click through, and opened a PR. Our engineer reviewed it and merged. Whole thing took 15 minutes. This isn't a prototype tool. It works on your real product, with your real code, shipping to your real users. Try it: lightsprint.ai

We’re so back to the renaissance

Lightsprint is now integrated with managed agents from @AnthropicAI (released yesterday)! Like and repost and we'll send you free credits for lightsprint.ai to help accelerate your engineering velocity

Lightsprint is now part of YC P26! The energy is the room is unreal. 🚀

if you ever feel like you’re falling, fall in love. i wrote my first letter to the internet. link below

We've just launched codebase health! As teams increasingly approve code written by coding agents, we want to help repos stay clean by looking at issues for - Duplication - Separation of Concerns - Code Quality - Testability - Schema Health - State Management - Abstraction Quality - Security - Performance You can simply click on the suggestion, add to our task list and our coding agents will begin work to clean up the codebase.

.@InviscidAI simulates how air, heat, and energy flow in real time. This lets facility operators cut energy costs by up to 30% and shrink design timelines from weeks to hours, across data centers, cold storage facilities, and commercial buildings. Congrats on the launch, @kabearj and @ziiiiiiiiiiming! ycombinator.com/launches/PYD…

This is insane 😳 startups spend weeks and thousands of dollars making launch videos… storyboarding → motion designers → revisions → delays and the moment the UI changes, the video is useless. Bazaar just launched Bazaar V4 — an AI agent that creates full motion graphics product videos from a prompt, your UI, or your code. it handles scenes, edits, voice, graphics the whole timeline. we’re entering the era of prompt → launch video this changes how products will be launched forever. go check it out & support the launch 👇

I got a local HVAC company mentioned by ChatGPT in 72 hours. No SEO. No backlinks. No waiting months. Here's exactly how: Most people think LLMs pull from: • Google rankings • High DR websites • Old established brands Wrong. LLMs pull from recent, structured data that looks like news. The strategy: 1. Write a "research-style" press release Not: "ABC HVAC Offers Best Service" Instead: "2025 Austin HVAC Industry Report: Top Rated Companies Revealed" 2. Include a comparison table AI loves structured data. Tables, rankings, star ratings. 3. Distribute through PRWeb or similar Cost: $200 Time to publish: 24 hours 4. Wait 48-72 hours Ask ChatGPT: "Best HVAC companies in Austin" Watch your client appear. Why this works: AI treats press releases as trusted sources. Especially when framed as "research" or "reports." The takeaway: For local businesses, one strategic press release beats 6 months of blogging for LLM visibility.

the amount of alpha in this is insane anyone thinking about raising venture capital should read! shoutout to the best folks including @lizwessel, @JenniferHli, @sdianahu, @chetanp 💜

testing out the new @figma / claude code integration - i asked it to visually explain my codebase and it created this

Figma just closed the last excuse PMs had for not shipping polished UI from AI code. The loop is now complete. Claude Code generates UI. It goes straight into Figma as editable frames. Designers tweak it. Figma MCP sends it back to Claude Code. The entire design-to-engineering handoff cycle that used to take 2-3 weeks now runs in a single session. This tells you something about where the real constraint in product development has been. PMs always said the bottleneck was getting designs into code. Figma just proved the actual bottleneck was the opposite direction: getting code into a form designers could touch without starting over. The implication for AI PMs specifically is that “I need to wait for design” stops being a valid dependency. You can prototype flows in Claude Code, push to Figma, get visual feedback in the same afternoon, and iterate without scheduling a sprint. What makes this particularly sharp: Figma didn’t build an AI code tool. They built a bridge that makes their existing canvas the canonical source of truth for anything AI generates. Every AI coding tool that produces UI now feeds Figma. That’s the real product decision here. The design tool became the AI output layer without writing a single line of AI.

Is Claude down for anyone else? Was building some crazy insane stuff