compression with a quality contract
A short course 3 modules · ~15 min total

Token Economics

Before you reach for a compression tool, understand the thing it compresses. This is a short, modular course — three focused lessons that take you from what a token is, through why agent loops get expensive, to how you compress them without changing a single decision the model makes.

The one-sentence version. An LLM charges by the token; an agent re-sends its whole history every turn, so cost grows with the square of the conversation; compression removes tokens the answer doesn't depend on; and the only honest compression is the kind you can prove didn't change the model's decision. This course builds that idea, piece by piece.

The three modules

What you'll be able to do

New here? Take the modules in order — each builds on the last. Already know tokens cold? Jump straight to Module 2 for the cost mechanics and the compression toolbox, or Module 3 for proof and the tier ladder. For implementation depth, the course cross-links to Concepts and Techniques.