Loading...

Tag trends are in beta. Feedback? Thoughts? Email me at [email protected]

Running Python code in a sandbox with MicroPython and WASM

Claude Fable is relentlessly proactive

Initial impressions of Claude Fable 5

GLM-5.2 is probably the most powerful text-only open weights LLM

Uber's $1,500/month AI limit is a useful signal for AI tool pricing

Datasette Apps: Host custom HTML applications inside Datasette

SQLite Does Not Accept Agentic Code

I think Anthropic and OpenAI have found product-market fit

Notes on Pope Leo XIV's Encyclical on AI

Vibe coding and agentic engineering are getting closer than I'd like

The last six months in LLMs in five minutes

DeepSeek V4 – almost on the frontier

The Zig project's rationale for their anti-AI contribution policy

Is Claude Code going to cost $100/month? Probably not—it’s all very confusing

The Axios supply chain attack used individually targeted social engineering

Changes in the system prompt between Claude Opus 4.6 and 4.7

Gemini 3.1 Flash TTS – with directed prompts

Qwen3.6-35B-A3B on my laptop drew me a better pelican than Claude Opus 4.7

Claude Token Counter, now with model comparisons

Anthropic's Project Glasswing sounds necessary to me

Profiling Hacker News users based on their comments

Pretext

Mr. Chatterbox is a Victorian-era ethically trained model

Thoughts on OpenAI acquiring Astral and uv/ruff/ty

Agentic Engineering Patterns

Package Managers Need to Cool Down

My fireside chat about agentic engineering at the Pragmatic Summit

AI should help us produce better code

Something is afoot in the land of Qwen

Clinejection — Compromising Cline’s Production Releases just by Prompting an Issue Triager

More →