AIAI Bytes
YOU
Module 2 of 7 Lessons coming soonStarting point ready

Prompt, Context & Structured Outputs

Can design a reliable prompt/context flow and diagnose whether a bad response is caused by instructions, context, or the model.

The lessons for this module are still being written. Its starting point is ready, so you can already choose how you'll start: with the code you wrote in Module 1, or with a clean copy.

Where do you want to start?

You haven't changed any Module 1 files in this browser yet. Either way, nothing you wrote is lost.

Continue with my codeDefault

Your Module 1 code moves into the Module 2 starting point. First we run your Module 1 tests, in your browser, to check it still works.

Start from a clean copy

Module 1 finished, with the reference code. Your own version is saved first, and your writing (Day 1 note, decision, ratings, write-up) comes with you.

What the check runs

Your Module 1 code tests: L1.1 Your first model call, L2.1 Log every call, L4.1 Write the assistant's rules, L4.2 Give it the policy. Your writing isn't tested here, and nothing calls a live model.

Detailspytest tests/m01 -m "not live and not submission"Checkpoint module-02-start · commit 14eea63
What's different in the starting point

Module 1's helper code that kept the app running before you'd built every piece is gone, because by now every piece exists. Module 2's own tasks are added when its lessons open.

DetailsFiles that differ from the finished Module 1: README.mdapp/assistant.pyapp/cli.py
← Back to Module 1