Week 01 — What is an Agent?

Lecture

• Slides (open in browser) — Course orientation · What is an agent? Press F for fullscreen, S for speaker view.

Lab

• Notebook: W1_lab_setup.ipynb (right-click → save) · read-only view
• Run it in Colab (course default): download the notebook, go to colab.research.google.com → Upload, and run the first cell — it installs dependencies and fetches the course library automatically.
• Prerequisite: an API key registered in Colab Secrets — follow the API Setup Guide before class.
• Goal: connect the model (“the brain”), observe that a call is text-in/text-out, write your first system prompt, and measure how it reduces fabricated answers on a trap-question set.

This week’s presentation

No paper presentation in Week 1 — presenter slots are assigned during orientation, and presentations begin in Week 2. Before your slot, read the presentation guide and study the sample deck: 6 slides maximum, 15-minute hard stop, cold-question Q&A.

Next week

W2 · Prompting & Reasoning — first student presentation: Chain-of-Thought (Wei et al., 2022, arXiv:2201.11903).