Prevent cognitive debt by manually retyping LLM-generated code

The true power of regular expressions (2012)

Octane – React’s programming model, compiled

The Shape of Things to Come

The AI Productivity Gap

AI migrated legacy COBOL programs to Java, bugs included

CP/M-386 – CP/M for 386 protected mode, derived from CP/M‑68K

The time filter in Google Search broke a few days ago

Running Kimi K3 on MI355X at Better Performance per Dollar Than B300

Go 1.27 Interactive Tour

Pgtestdb's template cloning approach to testing is fast

The Art of 64-bit Assembly

RipGrep musl binaries occasionally segfault during very-large searches

AI doesn't generate working products, that's still your job

Flint: A Visualization Language for the AI Era

Everyone is building LLM routers, we deprecated ours

qm

Run Kimi K3 using 29 GB of RAM at 0.50 tok/s

Google fixed more Chrome bugs in June than over the past two years, thanks to AI

DeepSeek-V4-Flash Update

The session you cannot take with you

Rune 1.1: adds Python, an Emacs editor, a symbol index and is now free

Agent Skill to Force Docs in ASD-STE100 Simplified Technical English

Stacked PRs are now live on GitHub

The Economic Benefit of Refactoring

OpenJDK Interim Policy on Generative AI

Why don't people use formal methods? (2019)

Go LLM SDK for streaming, tool-calling AI backends (plus frontend React lib)

GCC steering committee announces AI policy

You can't solve computer use by ignoring the interface

Agent-Manager: A Tmux TUI for Running Claude Code, Codex and OpenCode

Kuna: Decompiler Development in the Age of Coding Agents

LLM Honeypot

GitHub is the wrong shape for this new world

Kimi K3-256k

How to think about software quality (2022)

Launch HN: Tokenless (YC S26) – Automatic model switching to save money

Self-hosting Kimi K3: 20% more hardware cost, 20% better task resolution

TokenTown: A visual way to understand how LLMs work

Document-borne AI worms can self-propagate through Copilot for Word

Anthropic publishes a practical key-recovery attack on HAWK-256

Codex Security

Toolcraft

About the security content of macOS Tahoe 26.6

PyTorch: A Reference Language

"Opus 5 is a really bad model"

Using an open model feels surprisingly good

EYG: A Programming Language for Humans

Don't ask an LLM for a confidence score

My current strategy is to not read any of the code written by my agents