Talked to @willccbb turns out it’s really just what I thought it was. Agent with a submit for review function. Agent is jsut an llm in a form loop. Review function is just another agent with a rubric that approximates a loss function.↗

The AI SDK for TypeScript from Vercel, enabling developers to build AI-powered applications and agents with popular LLM providers. Includes support for generating UI and integrating with frameworks like React and Next.js.↗

1/ Base44 ended up making $189K *profit* in May - not $100K as expected. What changed: – Faster, unexpected growth – Drastically reduced LLM costs (see cost graph below) Sharing some details on what helped improve margins 🧵↗

Introducing apple-mcp - git.new/apple-mcp one simple command to give LLMs access to a bunch of apple-native tools like - contacts - notes - iMessages and more (soon) just add this to your claude desktop config. done.↗

i’ve been messing around with @duolingo a lot lately—not because i had some grand ambition to become fluent in another language, but because i thought it might impress someone (spoiler: she’s gone now). in my mind the most interesting thing though is what duolingo does for lang…↗

Have been playing with OpenAI's "Operator" agent all weekend. Some early thoughts: 1) The fact that it's already hosted and you can just start using it means it'll be more widely used than Claude's "Computer Use" feature. 2) Operator will be able to make websites that haven't…↗

Recently, I was at the library & stood in front of a book that might be a good read. I fired up Gemini, asked it for a plot summary & a digest of the reviews before deciding to read it. So I started a list : what are the ways I use AI that are new in the last year? I often publ…↗

LLM Fine tuning is here! San Francisco’s top AI engineers came together to see what’s possible with fine-turning and only 4 hours of hacking. Here’s an exclusive what we saw at the “Anything But Wrappers” hackathon (🧵):↗

To: My fellow data/info-biz nerds This is a great summary and another reason why proprietary quantitative or qualitative data & insights are going to be so powerful in this generative AI / LLM future h/t @joeobrien513↗

🗄️🗄️Will the next generation of LLMs have native data structures for storing insights? Pre-LLM era research has some interesting ideas here. Two of my favorite, both exactly what they sound like: - Stack-Augmented Recurrent NNs - Neural Turing Machine More 👇↗

piclaw is a self-hosted TypeScript project for building an OpenClaw-like AI coding agent with a web UI, using LLMs and supporting components such as Docker and VNC.↗

Pi coding agent extension that routes LLM calls through the Claude Code CLI.↗

t1code (T3Code in your terminal) provides a terminal-based TUI experience for coding with LLM/Codex-style workflows.↗