Stanford's new CS329Z course teaches building AI agents from scratch, from MCP to evals
stanfordnlp · x · 2026-10-31
Stanford is offering a new Fall 2026 course, CS329Z: Engineering AI Agents, taught by Diyi Yang and others, covering how to build, optimize, and evaluate working AI agent systems from scratch.
Topics include:
- LLM fundamentals, context engineering, and RAG
- Function calling, MCP, and tool use
- DSPy, LangGraph, and agent harnesses
- ReAct, agent memory, and multi-agent collaboration
- Prompt optimization, RL, and data engineering
- Agent evals, LLM-as-Judge, and safety mechanisms
- Coding agents and frontier applications
The course is hands-on: students build an agent harness from scratch, run agent evaluations, and deliver a full agent system as a final project. Course site, GitHub, and HW1 starter code are already public.
Related event: Stanford Launches CS329Z Course on Engineering AI Agents(3 posts)→
More from coding & agent
- ASC CLI now covers Xcode commands, fastlane migration, and Linux builds via xtool — rudrank · 2026-10-31
- LangChain founder: agent observability is 'Databricks for agents', not Datadog — hwchase17 · 2026-10-31
- A 710k-line PR ports pi-durable and Chord to Rust for crash-safe agent runs — jasonkneen · 2026-10-31
- Strix Halo one-command Docker for vLLM: Qwen3.8-Flash-Next hits 43 tok/s decode — asfbrz96 · 2026-10-31
- Anthropic's VS Code extension on Cursor backdoored for nearly 12 hours, up to 5,000 devs exposed — Ars Technica AI · 2026-10-31
- dhh on coding in the AI era: code is no longer the bottleneck, humans are — NERDDISCO · 2026-10-31