Writing · draft
What every team should know about LLMs
The fundamentals that stop cargo-culting — and make everything else make sense.
Draft in progress. The full piece is being written here. The seed below is the through-line; I'll expand it into the finished essay.
Tokens (models think in numbers, not words), the context window (why overstuffing it backfires), tools (why too many slow an agent down), and what an agent actually is — versus when you don't need one. Training, inference, and the difference between a transformer and a diffusion model, in plain terms.