The Complete LLM Engineering Course
This is the entire LLM Master course in one sitting — 20 chapters, front to back, for engineers who want the real mental model, not a highlight reel. We start from what a Transformer actually computes and don't stop until you've walked through agents, retrieval, production hardening, the major frameworks, vector databases, MCP, deployment, multimodal, the 2026 frontier, and running a capable model entirely on your own laptop.
Watch it straight through as a bootcamp, or jump by the chapter markers below and use it as a reference you come back to.
⏱ Chapters
0:00 Intro — how to use this course
1:32 Ch1 · Onboarding — what changed, your first API call, sampling knobs
9:17 Ch2 · Foundations — transformers, prompting, tools, structured output, caching, reasoning
23:02 Ch3 · Agent Engineering — ReAct, reflection, planning, memory, MCP, Skills
38:15 Ch4 · RAG & Knowledge — chunking, embeddings, hybrid search, eval, vector-DB internals
48:35 Ch5 · Production — eval, guardrails, security, cost, observability, orchestration
1:06:20 Ch6 · Frameworks — LangChain/LangGraph, AutoGen, CrewAI, LlamaIndex, computer use
1:17:16 Ch7 · ERP & ABAP for LLM engineers
1:27:20 Ch8 · Advanced/Expert — fine-tune vs RAG, quantization, LoRA, LLM-as-judge
1:40:40 Ch9 · Python foundations
1:49:26 Ch10 · Async, HTTP & DB foundations
1:58:19 Ch11 · FastAPI for LLM apps
2:07:56 Ch12 · LLM SDK comparison
2:19:04 Ch13 · LangGraph deep dive
2:27:42 Ch14 · Vector DB hands-on
2:36:07 Ch15 · MCP & tool-calling deep dive
2:44:28 Ch16 · Production deployment
2:52:06 Ch17 · Multimodal — vision, audio, realtime voice
3:00:45 Ch18 · Frontier 2025-2026
3:08:35 Ch19 · Four real-world projects
3:16:31 Ch20 · Gemma local mastery
3:25:48 Wrap-up
👉 Which chapter should I expand into its own standalone deep-dive next? Tell me in the comments.
#LLM #AIengineering #AIagents #RAG #PromptEngineering #MachineLearning #LLMops #MCP #FastAPI #VectorDatabase