Select Stage — Choosing What Goes Into the Context Window

~12 min read

Selecting (the book also calls this 'Reading') context means pulling previously-written information back INTO the context window when it's actually needed — from a tool, from memory, or from a knowledge base — based on relevance, recency, and importance.

Select Stage — Choosing What Goes Into the Context Window is a Pro topic

Sign in, then upgrade to Pro or Power to unlock this topic and the full AI Engineering curriculum.

Key points

  • Selecting (the book also calls it 'Reading') context means pulling saved information INTO the context window when it's actually needed
  • 3 sources: a tool (live call), memory (previously written facts), or a knowledge base (docs, vector DB)
  • Selection should be driven by relevance, recency, and importance — not 'include everything available'
  • Too little selected context starves the model; too much wastes tokens and can bury the genuinely relevant information in noise
  • This is the 'programming the RAM' half of the CPU/RAM analogy — deliberate selection, not indiscriminate inclusion