Saheb's AI Journey

It’s one thing to be aware of Sutton’s bitter lesson, quite another to internalise it; and even harder to resist encoding human bias, that we might think of as human knowledge into our agents.

An AI agent trained on a smaller board should be able to play well on a big board.

Every architectural layer is a compression operator.

Source: Kirk et al. 2023 ZSG survey (arXiv 2111.09794) + working sessions, July 2026. 1.

In the second part we talked about transformer heads and features.

In the first part, we discussed self play and the equilibrium our agent gets stuck in.

I had a lot of fun building an AI agent for Orbit Wars.

Building reinforcement learning environments is already a complex exercise in managing state, rewards, and training loops.

Reinforcement learning is interesting because it is intuitive and it aligns with how we learn naturally.