building @runtypelabs • 🤠→🌉 • 🧠 APIs, evals, carcinization

San Francisco
Joined January 2026
Nathan Booker retweeted
using opus 5.5 max fast to fix a typo
Long-range hornet nest removal by a kid
48
27
2
1,107
138,998
Nathan Booker retweeted
Great to see how teams are deploying Runtype at Agent Native Builders Hackathon at @Cloudflare HQ! Dozens of agents were built in 2 days in this room. Huge thanks to the teams who participated as well as the sponsors @Cotal_AI @MitosisLabs @nebiusai @tenki @immersivecommon
1
3
59
Nathan Booker retweeted
Introducing the open AI assistant for @WordPress Launching here at #WCUS It’s open source and powered by Persona.js, @RuntypeLabs OSS agent UI library for the open web And given we’re in PHX, here’s your gratuitous pool shot at 120 degrees 😎 Repo: github.com/runtypelabs/perso…
3
8
2
23
2,415
Nathan Booker retweeted
The context mentions and commands I was working through on persona.js are live 😎 Very helpful if you are building agent UI related to documents or code All of it's open source and plugable Check it: - persona-chat.dev/context-men… - persona-chat.dev/slash-comma…
1
1
71
Nathan Booker retweeted
Over the past 6 months we've maniacally prepped our repos to be coding agent friendly. Here are some things that worked. 1. Make CI blazing fast. Use every Rust, Zig, or Go ported tool that lets agents verify their work. This means UV, oxfmt, Typescript 7. Move integration tests to post merge hooks. 2. Trigger coding agents automatically based on triage labels. A coding agent should setup a proof of concept or repro steps automatically so an engineer can pick up the issue seamlessly. 3. Setup crons for things devs hate doing. Setup agents to fill SDK gaps, skill tuning, filling in critical regression checks. 4. Give agents ways to prove their work. Add screenshotting skills, agent-browswer, cloud storage for storing assets. 5. Make it possible to hermetically deploy your app, preferably multiple at a time. If a coding agent can deploy the app locally, the faster it can work. 6. Give the agents realistic production "simulation" data. Agents will work much better when they are working against data that looks like how your users use the product.
25
59
7
1,036
97,821
Nathan Booker retweeted
they don’t know fable 5 is back
296
377
43
9,349
325,274
Nathan Booker retweeted
Tested this one heavily before we released our implementation of it, mainly because I found it so useful. Really feels like a step change moment for open source AI. In one case it found a memory leak that 5.5 extra high and Opus 4.8 missed.
Kimi K2.7 Code is now live on Mixlayer. Frontier-grade agentic intelligence at a fraction of the cost: $0.75/1M input, $3.50/1M output. Moonshot's latest model significantly improves instruction compliance and long-horizon coding performance compared to K2.6 while also using far fewer tokens, 30% less on average!
1
1
2
243
Nathan Booker retweeted
1/ Today, I'm excited to announce ✨Mixlayer✨. Two years ago, I reoriented my life around a singular mission: to build the best inference platform for open source AI.
1
5
2
16
2,154