Baur Software
Insights on AI governance, automation, and technical strategy.
OpenAI published a giant Lean 4 repo claiming to solve a $1 million math problem. The headlines said the millennium mystery was finally cracked. So I read the thing. Here's what's actually in the box — and what the word 'sorry' means when a machine checks your homework.
Every agent framework treats every step the same way: send it to an LLM. You end up paying inference cost and latency for work a fifty-line function would do faster, cheaper, and deterministically. That's not a tooling gap — it's a missing primitive.
PAP ships eight agents that will never make an inference call. Not because they're simple, but because they don't need one. The work is structural: typed input goes in, a different typed shape comes out, deterministically, every time.
