Week 01 — Course Intro · What is an Agent?
| Material | Where |
|---|---|
| HTML chapter | Read online |
| Lab | Open in Colab · W1_lab_setup.ipynb download |
| Homework | No separate homework notebook in Week 1 |
| Papers | — (no paper presentation in Week 1) |
This week
After orientation, follow a fictional library-room reservation from the user’s goal through an availability check, a failed booking, and a confirmed alternative. Define an agent and distinguish the model, instructions, tools, and working state through their roles in that task. Compare a predefined workflow with model-directed decisions, then consider workflow patterns, the scope of autonomy, and evidence of completion.
Lab
W1_lab_setup.ipynb — first calls through aisuite; message roles, statelessness, temperature; and a final task: a system prompt that makes every answer come back as clean JSON, checked PASS/FAIL.
Click Open in Colab in the table above — the notebook opens directly in Colab. Save your own copy first (File → Save a copy in Drive) so your work persists; the .ipynb download stays available for local Jupyter. Then paste your API key into the setup cell and run the notebook top to bottom. Prerequisite: your own API key — follow the API Setup Guide before class.
Presentations
No paper presentation in Week 1. Presenter slots are assigned during orientation, and presentations begin in Week 2 with Chain-of-Thought (Wei et al., 2022 (PDF)) and Self-Consistency (Wang et al., 2022 (PDF)). Before your slot, read the presentation guide and study the sample deck linked there.