I analyzed hundreds of humans vs. AI Tetris games, here's what I found

Nice write-up! From the article:

> Instead of asking the model to choose a move, I asked it to write the logic that decides what a good move looks like. Once I reframed Tetris as a coding problem, everything changed.

Agreed. Nowadays the idea of invoking specialized MCP/skills is commonplace but OpenAI's ChatGPT was one of the first major LLMs to integrate with Python VMs that it could spin up as needed to workaround the inherent weaknesses of a language model.