AI & ML — Index
AI & ML — Reading Order
Read order: Karpathy Deep Dive (mental models, video) → LLM Frontier Learning Path (what next) → LLM Fundamentals (theory) → Langchain (framework) → MCP (protocol). Cross-domain synthesis (positioning this for AI-company interviews) lives at [[Learning/LLM & AI Stack]].
| Page | Summary |
|---|---|
| [[AI & ML/AI & ML Topics]] | Quick nav for AI/ML notes |
| [[AI & ML/Karpathy - Deep Dive into LLMs]] | Notes from Karpathy's 3.5h video: pretraining → SFT → RL/RLHF, hallucination fixes, tokens-to-think, Swiss-cheese capabilities, interview hooks |
| [[AI & ML/LLM Frontier Learning Path]] | Ordered videos/courses/posts after Karpathy: reasoning models, agents (loop, context engineering, multi-agent), LangGraph + GraphRAG, multimodal, Jev (TypeSafe System One model), CME295 + AskTGE-aligned shortlist with certifications |
| [[AI & ML/LLM Fundamentals]] | Transformer stack → LLM → training (LoRA, RLHF, quantization) → RAG, agents, diffusion |
| [[AI & ML/Langchain]] | Chains, agents, tools, memory, LCEL |
| [[AI & ML/MCP]] | Model Context Protocol — servers, tools, resources |
6 pages (+ lecture drawing [[AI & ML/Karpathy - LLM Intro.excalidraw]]). See also [[Interview Prep/System Design/Problem Designs/RAG & LLM System]] and [[Interview Prep/System Design/Problem Designs/ML Feature Store]] for the SD-interview angle on this material, plus the full GenAI SD track at [[Interview Prep/Content/System Design/GenAI/index]].