Explanation

Conceptual discussions that explain why things work the way they do.

Start here

These pages introduce the core ideas behind the framework, building from first principles to the complete system:

  1. The Environment Hypothesis – why AI output quality is an environment problem
  2. Context Engineering – teaching your AI what your team already knows
  3. Constraints and Enforcement – from good intentions to automated enforcement
  4. Codebase Entropy – why codebases rot and how to fight back
  5. Agent Orchestration – specialised agents with trust boundaries

  6. Compound Learning – how your AI gets smarter every session

  7. The Loops That Learn – four operational loops that make AI environments compound

Deep dives

These pages go deeper into the mechanics of each component:

Table of contents