Stanford CS329A publishes 9-lecture course on self-improving AI agents
le_james94 · x · 2026-09-16
- Stanford Online published the Autumn 2025 course CS329A "Self-Improving AI Agents" on YouTube, taught by Azalia Mirhoseini and Aakanksha Chowdhery (former tech lead of Google's 540B PaLM).
- Topics: self-improvement techniques (Constitutional AI, verifiers, test-time compute scaling, search + LLMs, RL train-time scaling), tool/code/memory augmentation, multi-step reasoning and planning, and agentic evaluation frameworks.
- Applications include coding agents, STEM research assistants, and autonomous robotics; the full paper reading list is public on the course site.
Related event: Three Weeks with Stanford CS329A: Generators Have Outrun Verifiers(9 posts)→
More from Companies & People
- Why OpenAI must copy Google's long game to keep up with Anthropic — haider1 · 2026-11-02
- US firms out-extracted Europe on IT via management — same may hold for AI — carlbfrey · 2026-11-02
- Sam Altman: AI query energy cost fell 1,000x-100,000x in 3-4 years — rohanpaul_ai · 2026-11-02
- Enterprise AI's real game: getting companies to spend far more, not less — inductionheads · 2026-11-02
- Sam Altman meets Samsung executives in Korea to deepen memory chip and AI infrastructure cooperation — pstAsiatech · 2026-11-02
- Tailscale called a 'single point of failure' for startups as acquisition chatter splits devs — TheZachMueller · 2026-11-02