Python AI Assistant camp plan
Use this page when running Python AI: Build a Smart Assistant as a 5-day camp.
At a glance
| Day | Blocks | Goal |
|---|---|---|
| 1 | Setup, Stage 1, Stage 2 | Local Python works and the assistant greets the user by name. |
| 2 | Stage 3, Stage 4, Stage 5 | The assistant reacts, keeps listening, and uses functions. |
| 3 | Stage 6, Stage 7, Stage 8 | Students add jokes, modules, and keyword understanding. |
| 4 | Stage 9, Stage 10, catch-up | The assistant remembers data and connects to a live API. |
| 5 | Bug fixes, personality polish, parent demo | Students show commands and explain the rule-based logic. |
Minimum viable finish
A student should leave with one runnable assistant.py that starts with uv run, answers typed input, and includes at least one custom response.
Coach triage
- If local setup fails, pair the student with a working machine for the lesson and fix setup during break.
- If imports fail, run
uv add rich requestsfrom the project folder. - If speech fails, keep typed output working; the
saycommand is a bonus, not the core concept. - If a student is ahead, have them add a new command family with clear keyword matching.
Common stuck points
- Students run Python from the wrong folder.
- A
whileloop may not have a clean quit path. - The live joke/API step needs an offline fallback to keep demos reliable.