Lobster Roll

All Programming (General) 40073 Systems / Low-Level / OS 27683 Programming Languages / CS Theory 20658 Data / Databases / Infrastructure 15947 Web Development 13873 Culture / Philosophy / History / Reading 12379 Security / Privacy 12212 Productivity / Career / Business 9226 Maker / DIY / Hardware 6758 AI / Machine Learning 5030 Science / Math / Physics 4886 Design / UX / Visualization 4401 Gaming / Retro Computing 3982 Apple / macOS / iOS 3142 Internet / Digital Culture 1581 Crypto / Blockchain 427 Health / Fitness / Lifestyle 109

AI / Machine Learning

CONTRACT.md: The Naughty List for AI Coding Agents (discussdontcode.com)
Reimplementing FlashAttention for performance and giggles (aminediro.com)
Calibre adds AI 'discussion' feature (lwn.net)
Maestro - Desktop AI Agent Command Center (Free/OSS) (runmaestro.ai)
Rob Pike Goes Nuclear over GenAI (imgur.com)
Toward a policy for machine-learning tools in kernel development (lwn.net)
Using Vectorize to build a search engine in 160 loc (blog.partykit.io)
Advent of Code 2025: The AI LLM Edition (github.com)
How and why I moved from Apple Services to my own server (bastiangruber.ca)
"AI" is bad UX (buttondown.com)
Attention From First Principles (metaworld.me)
We asked four AI coding agents to rebuild Minesweeper—the results were explosive (arstechnica.com)
An Appeal from the United Federation of Dictators, Despots, Evil Emperors and Tyrants (nibblestew.blogspot.com)
How I taught an AI to use a computer (blog.jamesmurdza.com)
Galois Slicing as Automatic Differentiation (arxiv.org)
Thirteen years of Rust and the birth of Rue (steveklabnik.com)
How I protect my forgejo instance from AI Web Crawlers (her.esy.fun)
Task Injection – Exploiting agency of autonomous AI agents (bughunters.google.com)
Text similarity search via normalized compression distance (discourse.haskell.org)
Updated LLM Benchmark (Gemini 3 Flash) (entropicthoughts.com)
Porting an HTML5 Parser to Swift (ikyle.me)
Porting a complete HTML5 parser and browser test suite [from Python to OCaml using LLMs] (anil.recoil.org)
Prompt caching: 10x cheaper LLM tokens, but how? (ngrok.com)
My first win building with agents (olano.dev)
dogalog: Prolog-based livecoding music environment (github.com)
Toad is a unified experience for AI in the terminal (willmcgugan.github.io)
Your job is to deliver code you have proven to work (simonwillison.net)
Dear ACM, you're doing AI wrong but you can still get it right (anil.recoil.org)
Microsoft quietly kills IntelliCode in favor of the paid Copilot (visualstudiomagazine.com)
lightning-extra: PyTorch Lightning plugins and utilities for cloud-native machine learning (github.com)