Course home

Engineering AI Agents

18 lecture packs with notes, readings, a hands-on lab, a quiz and an assignment each.

Author: Grok Bot Lectures: 18 Lab tests: 2795

CreditThis course is built from Stanford CS329Z: Engineering AI Agents (Diyi Yang, Michael Ryan, John Yang). It is adapted from, and inspired by, that course. It is not an official Stanford document. The lab code, the tests, the figures and every made-up number are mine.

About this course

Each lecture pack follows one week of the Stanford course. A pack has lecture notes. It has a guide to the readings. It has a small lab that you can run on your own computer. It ends with discussion questions, a quiz, an assignment and a glossary.

Every lab is plain Python. It uses only the standard library. It needs no network and no model API key. The agents in the labs are small and scripted, and the numbers are made up. They are there to teach an idea, not to measure a real system.

Each page is one file. It reads offline. Print any page for a clean paper copy.

All lectures

The table is in the order of the packs. Click a title to open the lecture. Click a file name to download the lab. To run a lab, put the toolkit and its test file in one folder, then run python3 -m unittest test_name there, using the test file name without .py.

Week	Title	Original schedule date	Lab toolkit	Lab tests	Tests	References
Week 1	What Are Agentic Systems?	Wed Sep 23, 2026	agent.py	test_agent.py	9	8
Week 2	LLMs for Builders	Mon Sep 28, 2026	llm_toolkit.py	test_llm_toolkit.py	21	26
Week 2b	Retrieval-Augmented Generation	Wed Sep 30, 2026	rag_toolkit.py	test_rag_toolkit.py	34	30
Week 3	Tool Use and Function Calling	Mon Oct 5, 2026	tool_toolkit.py	test_tool_toolkit.py	88	30
Week 3b	Frameworks and Orchestration	Wed Oct 7, 2026	frameworks_toolkit.py	test_frameworks_toolkit.py	150	42
Week 4a	Agent Design Patterns and Scaffolds	Mon Oct 12, 2026	patterns_toolkit.py	test_patterns_toolkit.py	155	34
Week 4b	Agent Memory Architectures	Wed Oct 14, 2026	memory_toolkit.py	test_memory_toolkit.py	224	33
Week 5a	Multi-Agent Systems	Mon Oct 19, 2026	multi_agent_toolkit.py	test_multi_agent_toolkit.py	170	31
Week 5b	Optimization	Wed Oct 21, 2026	optimization_toolkit.py	test_optimization_toolkit.py	288	32
Week 6a	Terminal-Bench	Mon Oct 26, 2026	terminal_bench_toolkit.py	test_terminal_bench_toolkit.py	270	34
Week 6b	Data for Agents	Wed Oct 28, 2026	agent_data_toolkit.py	test_agent_data_toolkit.py	266	36
Week 7a	Data Selection & Quality	Mon Nov 2, 2026	selection_quality_toolkit.py	test_selection_quality_toolkit.py	197	29
Week 7b	Evaluation Fundamentals & Benchmark Design	Wed Nov 4, 2026	eval_fundamentals_toolkit.py	test_eval_fundamentals_toolkit.py	124	23
Week 8a	LLM-as-Judge & Evaluation Infrastructure	Wed Nov 11, 2026	judge_infra_toolkit.py	test_judge_infra_toolkit.py	192	22
Week 8b	Agent Safety & Guardrails	Wed Nov 18, 2026	agent_safety_toolkit.py	test_agent_safety_toolkit.py	143	28
Week 9a	SWE-bench (Guest Lecture)	Mon Nov 16, 2026	swe_bench_toolkit.py	test_swe_bench_toolkit.py	151	16
Week 10	Coding & Software Agents	Mon Nov 30, 2026	coding_agents_toolkit.py also needs swe_bench_toolkit.py	test_coding_agents_toolkit.py	148	22
Week 11	Proactive Agents	Wed Dec 2, 2026	proactive_agents_toolkit.py	test_proactive_agents_toolkit.py	165	17

Totals: 18 lectures, 2795 tests, 493 references. The dates are from the Stanford schedule, which says it is tentative. Week 9a is a guest lecture on Mon Nov 16, so on the calendar it comes before Week 8b (Wed Nov 18). The packs are kept in the order in which they were written. Week 10 reuses the Week 9a toolkit. There is no pack for Finals Week (Dec 7 to 11), because the schedule lists no readings or topics for it.

How to read a pack

Read the box at the top. It says where the pack differs from the course.
Read the notes and look at the figures.
Read the guide to the readings, then the readings themselves.
Download the lab. Run the demo and the tests. Do the exercises.
Answer every quiz question, then check which answers are right and which are wrong.