The Only Two Courses You Need for Modern Deep Learning: Stanford CS312 and CS336
ChengleiSi · x · 2026-11-19
An X user recommends just two Stanford courses for learning modern deep learning / LLMs:
- CS312: Deep Learning Alchemy (Fall 2026, Tatsunori Hashimoto et al.): builds empirical intuition via heavy experimentation and outcome-prediction quizzes; 5 units, no grade inflation; all materials public
- CS336: Language Modeling from Scratch (Spring 2026, Percy Liang et al.): build a full LLM from data cleaning and pretraining to evaluation, with far more code than typical courses and a focus on multi-GPU efficiency; recordings on YouTube
Both are hands-on, experiment-driven, and suited to learners with Python and math backgrounds who want deep understanding.
More from Companies & People
- Google ships Genie 4, Gemini 4 Neon and Stitch Loop; Nvidia falls despite earnings beat — xiaohu · 2026-11-19
- OpenAI cofounder at the White House: agent-driven cyberattacks are a top infrastructure risk — ibab · 2026-11-19
- OpenAI blog details how AI is reshaping its own R&D process — daveholtz · 2026-11-19
- Ex-NVIDIA employee: Tesla and SpaceX feel culturally like early NVIDIA — yunta_tsai · 2026-11-19
- Travis Kalanick warns mob mentality can break founders — even Jensen Huang — edgarpavlovsky · 2026-11-19
- Skeptic questions Anthropic's claim of bigger AI speedups than OpenAI's — ziv_ravid · 2026-11-19