Ala BEN KHALIFA
All AI sessions

Prompt Cache Lab

Session 3 recap · Pricing & Caching · Sat 17 Oct

Every request re-sends the whole context; caching makes the repeated part cheap. The one-page recap: what you pay for, how caching cuts it, five habits, two myths, and a prompt to keep.

Download the PDFOpen the image (opens in a new tab)
The session 3 recap sheet, Prompt Cache Lab: what you pay for in tokens, how caching makes the repeated part cheaper, five habits for your next chat, two myth busters, and a prompt to keep.Open the full-size image (opens in a new tab)

A prompt to keep

Stable parts first, today's task last, so the next request can reuse the cached start.

[Stable: paste AGENTS.md team rules here]
[Stable: paste checkout-spec.pdf discount rules here]
Today's task: write test cases for stacked coupons.

Sources

Everything on the sheet is checked against these.