1. Preface
2. Setting Up the Environment
5. 1. Week 1: From Matmul to Text
   1. 1.1. Attention and Multi-Head Attention
   2. 1.2. Positional Encodings and RoPE
   3. 1.3. Grouped/Multi Query Attention
   4. 1.4. RMSNorm and MLP
   5. 1.5. The Qwen3 Model
   6. 1.6. Generating the Response
   7. 1.7. Sampling and Preparing for Week 2
6. 2. Week 2: Tiny vLLM
   1. 2.1. Key-Value Cache
   2. 2.2. Quantized Matmul (2 Days)
   3. 2.3. Flash Attention (2 Days)
   4. 2.4. Continuous Batching (2 Days)
7. 3. Week 3: Serving
   1. 3.1. Paged Attention, Part 1
   2. 3.2. Paged Attention, Part 2
   3. 3.3. Mixture of Experts
10. Glossary Index