Module 1Beginner
LLM foundations
Covers LLM inference loop, Self-attention, Prefill vs decode, KV-cache, Tokenization.
Revise in a page
Cheat sheets
11 one-page summaries, one per module. Read one on screen to revise, or print it and keep it by the desk.
Module 2Beginner
Prompting and structured output
Covers Prompting types, Structured output options, Reasoning models, Structured output, Prompts.
Module 3Beginner
Embeddings and vector search
Covers Embeddings, Where it fails, ANN indexes, Choosing a store, Production.
Module 4Intermediate
Retrieval-augmented generation
Covers RAG pipeline, Hallucination mitigation, RAG evaluation, Interview traps, Locate a failure.
Module 5Intermediate
Multimodal and document AI
Covers Images cost tokens, Four ways to read a page, Parsing caps quality, Isolate the layer, Chunking parsed documents.
Module 6Advanced
Tools and agents
Covers Tool/function calling, Agent = LLM in a loop, Planning, Tool calling, ReAct.
Module 7Advanced
Context, memory and multi-agent systems
Covers Context engineering, Memory, Context failures (course taxonomy), Agent design, Agent evals.
Module 8Advanced
Safety and guardrails
Covers Prompt injection, Layers (defence in depth), Limit the blast radius, Interview traps, Injection defence.
Module 9Production
Production: cost, serving and observability
Covers Formulas, Cost levers, Latency levers, Observability, Measure first.
Module 10Production
Evaluation and model adaptation
Covers Prompting vs RAG vs fine-tuning, Prompting, RAG, Fine-tuning, Fine-tuning methods.
Module 11Production
GenAI system design
Covers The method (nine links, in order), RAG QA system, Agent product, Content moderation, Always volunteer in a GenAI design answer.