An AI agent is just a while loop. I built one in 70 lines of Python, then tricked it into leaking my .env

Temperature 0 isn't deterministic. Here's why the same prompt returns different text, and how to stop your LLM tests from flaking.

I'm a software engineer based in Japan. The progress of LLMs and tools like Codex and Claude has been...

Executive summary Something happened in July 2026 that has not yet been absorbed by the people who...

Most AI products start on a frontier API and stay there until the bill, the latency or legal forces a...

How do you regression-test a system that can answer differently twice? Part 8 findings of an...

Back in June I wrote that AI coding agents need project memory, not just bigger prompts. The idea was...

I rebuilt my tiny homage to Warhammer 40k this week. The original, Hive City Rampage, was a...

Why ? I’ve always been interested in how we use language technologies, especially when...

Every AI-and-engineering post right now is about adding a guardrail. Lint rules the agent can't...

We trained a text-to-speech model with voice cloning that fits into 100 MB (fp16) and runs on-device...