Nathan Lambert says his RLHF book is helping readers land jobs at AI labs
natolambert · x · 2026-11-06
Nathan Lambert, author of Reinforcement Learning from Human Feedback / Post-Training, shares that readers regularly tell him the book helped them land jobs at AI labs by building real intuition for how training methods work. The free web book covers the full post-training pipeline — instruction tuning, reward modeling, rejection sampling, RL, on-policy distillation, direct alignment — plus advanced topics like synthetic data, tool-use, character training and evaluation, with companion code, tooling and a course. He notes idea/math-heavy writing doesn't sell many books, but rewards those who put in the effort.
More from Companies & People
- 7-person team Krea ranks #3 in Multi-Image Edit Arena, CEO discusses training and open-source plans — arena · 2026-11-06
- CFO: We Have 20 AI Use Cases We Can't Execute — Four Questions to Cut the List — alex_verem · 2026-11-06
- SF Systems Reading Group (SFSRG) named after 15 sessions at Greylock — _shreya_s · 2026-11-06
- Stealth startup's Cortex scores 73.2% on τ³-bench banking, beating frontier LLMs by 18 points — multiply_matrix · 2026-11-06
- Isomorphic Labs to present AI drug design philosophy at NeurIPS 2026 — EricBuess · 2026-11-06
- Epic CEO Tim Sweeney: 'AI is happening, AI won't un-happen' — Epic goes all-in on AI for game devs — mark_k · 2026-11-06