Tracks — guided reading paths

Staged, ordered routes that tell you what to read, in what order, to master a topic.

• Post-training LLMs — How a base model becomes a frontier assistant — RLHF, preference optimization, and RL for reasoning.
• The Transformer, end to end — From the attention mechanism to modern architectures — how today's models actually compute.
• Serving LLMs efficiently — Make a trained model fast and cheap to run — memory, batching, speculation, disaggregation.
• Training at scale — Spread one model across thousands of chips — the parallelism stack behind frontier training.
• Diffusion & generative vision — How models learn to generate images — from denoising to flow matching to modern text-to-image.
• The DeepSeek lineage — One lab’s stack, paper by paper — GRPO, MLA, MoE, and pure-RL reasoning.
• Reasoning & agents — From a single prompt trick to verifier-checked, tool-using agents — how models learn to think.
• Open model architectures — How the frontier open models are actually built — the design choices, paper by paper.
• Vision & multimodal — How models learned to see, hear, and connect images to language — and to generate them.
• Deep learning foundations — The bedrock under everything else — optimization, depth, attention, scale, and adaptation.