The Lost Atomic Update on Loongson CPU

Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint

How to keep enjoying programming in a world of LLMs

Leaving them behind

One Month Without AI

I asked Meta’s Muse for its filesystem and it sent me 6.8GB

Measuring the sloppiness of code

Debian Code Search: Fast TurboPFor with Go SIMD

MicroLLM Lab – Try 7 tiny LLM's in the browser

AI Workers' Inquiry 2026

Parley: Federated, decentralised chat that speaks plain IRC

Replacing Pull Requests with Delta

Robot-use agents

Turbo Haskell

The Normalization of Inexplicable Failures

We Should Be Able to Change Our Languages

Automatically detecting AI text in my browser

Useful Things Agents Can Do That Are Not Writing Code

A heap overflow and SSO misconfiguration to compromise OpenAI internal repos

The Shape of Things to Come, Part 1: The Continuous Thunderdome

AI Has No Wisdom and Neither Will You

Check if a file was made with Claude

Writing Rust code that's fast by asking agents to make the code faster

GPT-6 Astra Solves a WWI German Radio Cipher

Vale, code-like linting for prose

Wrong, not broken

Plain-text files are at risk

kicking the tires on jev (TypeSafe's System One model) with 2048

There's no point at which turning your brain off will work

How OpenAI Used Its Own LLMs to Design Its Jalapeño Chip

More →