Learn
Short, interactive lessons on AI concepts. Each one is a stack of small widgets you can poke - the goal is intuition, not coverage.
Read the prose, move the slider, watch what changes. These pages are meant to be touched, not just read.
Lessons
How an agent works
An agent is a model in a loop. Step through one turn at a time, watch the context grow, and see why "agent" is a runtime pattern, not a kind of model.
How an LLM works
An LLM predicts the next token. That's the whole game. Step through tokenization, embedding, parallel prefill, the KV cache, and sequential generation.