We work in the open: every experiment here follows the same loop — question, action, evidence, decision, next evidence. These write-ups are where that loop gets told properly, including the runs that didn't work.
Baseline retrieval found the expected package in the top 10 for 62% of task queries. LLM enrichment moved the number exactly nowhere. The full evaluation story — and why 83% was a number we refused to ship.
Read ▸ September 5, 2026GitHub's Advisory Database teaches an agent to avoid packages already known to be dangerous. It cannot tell an agent that an unknown package is trustworthy. Notes from a conversation about the difference.
Read ▸ September 2026Real geodesy in the browser: NumPy running in Pyodide, a Three.js globe, Natural Earth geometry, and distortion circles after Tissot's 1858 idea. How Maps with Python came together.
Read ▸ September 2026Five clues per package, ordered as a single information gradient, with the modal solve landing at clue two or three. How we fact-check every rung — and decide which famous stories not to use.
Read ▸Why we think an agent that executes a small step, observes the real result, and leaves behind a replayable notebook beats one that just answers.
Drafting…The operating formula behind every run of this team: why prototypes stay subordinate to decisions, and what changes when negative results are published first-class.
Drafting…