Syllabus

Date	Topics	Video
September 25^th, 2026	Lecture 1: Transformers • Background on NLP and tasks • Tokenization • Embeddings • Word2vec, RNN, LSTM • Attention mechanism • Transformer architecture • End-to-end example	Coming soon, stay tuned!
October 2^nd, 2026	Lecture 2: Large Language Models • Transformer model families • LLM definition and architecture • Mixture of experts • MHA, MQA, GQA • Position embeddings (RoPE and variants) • Context length, temperature • Sampling strategies	Coming soon, stay tuned!
October 9^th, 2026	Lecture 3: LLM training • Pretraining • Supervised finetuning (SFT) • Parameter-efficient finetuning (LoRA) • Preference tuning (RLHF, DPO) • Reasoning • On-policy distillation (OPD and variants) • Distillation to smaller models	Coming soon, stay tuned!
October 16^th, 2026	Lecture 4: Reinforcement learning with LLMs • Mathematical conventions • Reward design • Policy gradients • Limitations • Preference tuning with PPO (RLHF) • Reasoning with GRPO (RLVR) • On-policy distillation	Coming soon, stay tuned!
October 23^rd, 2026	Midterm
October 30^th, 2026	Lecture 5: LLM systems • Distributed training • Inference optimizations • KV caching • Speculative decoding • Efficient kernels • Flash Attention • Hardware trade-offs	Coming soon, stay tuned!
November 6^th, 2026	Lecture 6: AI Agents • Tool calling • MCP • Memory, retrieval • Context compaction • Harness optimization • Coding agents • Skills, plugins	Coming soon, stay tuned!
November 13^th, 2026	Lecture 7: LLM evaluation • LLM-as-a-judge overview • Best practices and benefits • Biases and pitfalls • Agent evaluation • Benchmarks	Coming soon, stay tuned!
November 20^th, 2026	Lecture 8: Diffusion LLMs • Continuous diffusion • Discrete diffusion • Masked diffusion • Training • Inference	Coming soon, stay tuned!
December 4^th, 2026	Lecture 9: Trending topics • Recap • Multimodality • Closing thoughts	Coming soon, stay tuned!
December 9^th, 2026	Final