AI by Hand ✍️

AI by Hand ✍️

Retrieval per Turn

Paying for the search every turn

Prof. Tom Yeh's avatar
Prof. Tom Yeh
Aug 19, 2026
∙ Paid

Library › Context Problems

  1. The Context Window

  2. Window Occupancy

  3. Growth per Turn

  4. Remaining Space

  5. Turn Budget

  6. Retrieval Footprint

  7. Retrieval per Turn

  8. Search Rounds

  9. Window Sizing

  10. Truncation

  11. Compaction

  12. The Compaction Bill

  13. Compacted Retrieval

  14. The Compaction Threshold

  15. Verbatim Tail

  16. Stateless API

  17. Retrieval Cost

  18. Context Tax

  19. Quadratic Cost

  20. Cost Forecast

An agent that retrieves on every turn pays for the retrieval on every turn. The document does not replace the last one, it joins it: each turn adds its own chunks to the window alongside the question and the answer, so a turn costs far more than the words anyone typed. At two tokens a word, even a short document doubles what a turn costs, and the conversation around it is the smaller part. That is why an agent with retrieval runs out of window in a handful of turns while a plain chat runs for dozens.

Paid members: the worksheet and its printable PDF are below ↓

This post is for paid subscribers

Already a paid subscriber? Sign in
© 2026 Tom Yeh · Privacy ∙ Terms ∙ Collection notice
Start your SubstackGet the app
Substack is the home for great culture