wanghao-SunSky/cs329z-chinese

What is this?

A free, Chinese-language study guide for Stanford's CS329Z course ("Engineering AI Agents"). It's a self-paced tutorial for Python developers who want to learn how to build, evaluate, and improve AI agents — programs that use large language models (LLMs) to perform tasks. The guide covers topics like retrieval-augmented generation (RAG), tool calling, memory, multi-agent collaboration, and safety.

It's not an official Stanford product — it's an independent, community-made resource based on the course's public materials.

What's inside?

• 19 lessons in Chinese, starting with agent basics and moving to advanced topics.
• 4 hands-on exercises — for example, building an agent from scratch, designing evaluation methods, working on a course project, and practicing with research papers.
• A code-reading guide for the course's homework 1 and lesson 5 companion code.
• HTML versions of the whole guide — a searchable website with a table of contents and light/dark themes, plus a single-file offline version.
• A course guide that maps out a suggested learning path.

How to use it

• Read online via the provided link, or download the repository and open html/index.html in a browser. For offline use, open html/offline.html — it's a single file.
• Recommended path: read lessons 1–4 first, then do homework 1 (HW1) to get hands-on.
• If you modify the Markdown source and want to rebuild the HTML, you'll need Node.js 18+ and Python 3, then run the two commands shown in the README.

Who is it for?

Python developers who already know basic API calls and want a structured way to learn agent development. No prior agent experience is required.

Feedback and corrections

The project welcomes bug reports and suggestions via GitHub Issues — just mention the chapter and specific paragraph.