All ideasMore in Work
Work

I let Astra run a software factory for 35 hours. It built nothing of value.

Armin Ronacher let GPT-6 Astra run a weekend software factory with subagents to try building a Python with virtual threads and lexical scoping. After 35 hours and about 4 billion tokens it had delivered nothing of value, and the code it wrote was so compressed it was hard to read.

This one went wrong. Read what happened so you don’t hit the same problem.

How it works

  1. Sets up a factory where Astra manages its own context, keeps notes in an agent-notes folder, and spins off subagents
  2. The factory works for 35 hours on a Python variant
  3. It produces a lot of code but nothing of value, in a compressed style that hurts readability

Why it’s worth knowing

The model was trained to finish tasks and save tokens, so it wrote code squeezed so tight that the whole codebase became hard for a person to read.

Where it’s from

Summarized from a public post on lucumr.pocoo.org. See the original ↗

Upgraded prompt

A ready-to-use prompt for this idea, built from research on what makes AI agents do the job right: who the AI should be, the exact steps, and a check to make sure it was done the way you wanted.

1Who the AI should be
2The prompt, with the steps
3The check when it’s done
Upgraded prompts are coming soon.

Try next

WorkA contractor talks measurements into his phone and QuickBooks sends the estimate in 45 secondsWorkThe AI audited my eBay store, found 28 listings going nowhere, and sold a bundle for meWorkAn agent now makes the promo posts for our 70 radio shows a monthWorkOne man released a whole movie with about 40 tools he built with AI

Every idea is a real post from someone who tried it, summarized in our words, with a link to the original. App names are swapped for plain words.