Stanford’s 2025 CS329A course focuses on self-improving AI agents
caglar_ee · x · 2026-08-04
Stanford’s CS329A course, Self-Improving AI Agents, is being taught in Autumn 2025 by Aakanksha Chowdhery and Azalia Mirhoseini.
The syllabus spans:
- self-improvement methods for LLMs, including constitutional AI, verifiers, search with LLMs, and test-time compute / RL scaling
- tool use, code, memory, and multimodal orchestration for agents
- multi-step reasoning, planning, and robust evaluation frameworks
- applications in coding agents, STEM research assistants, and autonomous robotics
The course is project-based and centers on papers, readings, and invited speakers from academia and industry.
Related event: Stanford Launches Self-Improving AI Agents Course(4 posts)→
More from coding & agent
- DeepVoyager-VL: Building Long-Horizon Multimodal Agents Without RL — Huanyao Zhang · 2026-08-04
- Essential Linux Server Hardening Guide for Vibecoding Beginners — lxfater · 2026-08-04
- Pain Point: Why Agent Harnesses Constantly Need Updates for New Models — EdwardCHWang · 2026-08-04
- Qwen 3.8 Max Tested: Rivals Frontier Models in Coding and Multimodal — WorldofAI · 2026-08-04
- Dev Builds Free 'Radar' Scanner to Catch Security Gaps in AI-Generated Websites — indishere · 2026-08-04
- Include Standard Libraries in RL Environments to Stop AI Agents Reinventing the Wheel — cephaloform · 2026-08-04