The Ant Lab - understand agents by building them

Build one simulation loop and swap four different brains into it - hardcode, Q-learning, LLM, LLM+memory. That is the mental map for every agent system: the environment is decoupled from the decider, and “AI” is the swappable box inside a loop that existed before it.

Chapters

1. 01The world as a model5 min
2. 02The one idea6 min
3. 03The world12 min
4. 04Marks & communication8 min
5. 05Hardcode8 min
6. 06Q-learning14 min
7. 07An LLM steers12 min
8. 08Memory & skills10 min
9. 09Running it as a workshop6 min