bookmarks
3,996 bookmarks

Running a tight group of "hyper engineers" who spend $2,000-$20,000 per month on personal codegen. Usually clearing >10B tokens per year. (Some real lunatics) We share a bunch of spending + managemrnt hacks If you feel like this is you, shoot me a DM of your spend receipts↗

Taken Sep 1, 2016 when @JonathanRoss321 convinced me we could take on the giants, build new silicon and that AI was coming. In typical SV fashion, we didn’t even have a company yet - just a term sheet from me to invest and the three of us. I spent the next month recruiting as man…↗

·@chamath·Dec 24, 2025·essay·ai·groq·nvidia·silicon

At @every, we created our 2026 strategic plan by building a chatbot that asks you questions and debates your answers. It decreased our planning cycle time by 60% and improved thoroughness and alignment by 3x, easily. The chatbot had access to our top-level strategic plan and m…↗

·@bran_don_gell·Dec 24, 2025·repo·ai·chatbot·claude·notion

Claude Code is the single most powerful AI tool ever created It allows ANYONE to build the app of their dreams. No coding experience required Looking to master it? Here is every tutorial video I put out in 2025: 🧵↗

·@AlexFinn·Dec 24, 2025·tutorial·ai·app-development·claude-code·no-code

Because of LLMs, the only type of people I’d hire now are what I’d call “super-unicorns” - Developer - Designer (or great design taste) - Product-focused (vs feature-focused) - Marketer - Agent expert - High emotional intelligence - 5+ years experience creating web app products…↗

I gave Claude full access to the / directory of a Linux virtual machine. I then asked it to create a "child" Claude that it can talk to and do anything together. I gave it a domain and it figured everything out on its own. You can watch the stream here: clopus.live↗

·@kuberdenis·Dec 18, 2025·demo·ai·automation·claude·linux

Introducing Vercel Knowledge Base. Guides, tutorials, and 🔜, videos and demos, for developers building on Vercel. We created it so anyone at @vercel can share their expertise. This collective knowledge is used to power our support agent, AI chat, and more.↗

OpenTUI is really incredible and its core is written in zig. Between that and libghostty we really have the makings of a common foundation for tui applications. I’ve now played doom and watched a move in my terminal and those are just delightful side effects of having a proper ab…↗

·@tobi·Dec 15, 2025·repo·js-ts·libghostty·opentui·tui

Unpopular opinion: Current code review tools just don’t make much sense for AI-generated code When reviewing code I really want to know: - The prompt made by the dev - What corrections the other dev made to the code - Clear marking of code AI-generated not changed by a human↗

There are no limits anymore. Anyone can do anything. The only limiting factors are agency and ambition. Never has a college degree, work experience, network, even the accumulation of knowledge been worth less. You can just ship things.↗

·@rauchg·Dec 13, 2025·essay·agency·ambition·career·motivation

It changed mine too. For some reason I can now yell by typing on my keyboard.↗

·@badlogicgames·Dec 12, 2025·media

I'm thinking about this a lot lately — did we only ever create abstractions to avoid duplication, and now is maintaining duplication actually not really a problem? I'm working on this very componentized marketing site kit product we're building and can't shake the feeling that ma…↗

Agentic coding did change my brain in terms how I think about code - especially the terms. It *weaves* features in, it's all *shapes* of code. It *traces* through it. What other terms did you learn?↗

Most of coding was never about writing code. AI is just making this more obvious. You no longer need to recall syntax, function structure, boilerplate code, or even API endpoints. That’s the easy part and AI is very good at it. The hard part was never typing. It was always th…↗

·@Prathkum·Dec 9, 2025·essay·ai·coding·programming·thought-process

Don't think of LLMs as entities but as simulators. For example, when exploring a topic, don't ask: "What do you think about xyz"? There is no "you". Next time try: "What would be a good group of people to explore xyz? What would they say?" The LLM can channel/simulate many pe…↗

·@karpathy·Dec 7, 2025·essay·ai·andrej-karpathy·language-models·llm

@steipete Have you played with their tunnels yet? They beat ngrok in my opinion and you can their APIs to auto provision new ones dynamically. That’s what I do for new feature branches of my web apps. Nuts that it’s all free too!↗

·@rida·Dec 6, 2025·essay·api·free·ngrok·tunnels