📄️ Table of contents
Free discovery course on large language models in 6 lessons. What they are, how they are trained in three stages, what RAG solves, how agents and tool use work, and how to choose and secure one.
📄️ 1. What makes a model large
Lesson 1 of the free LLM course: what parameters are, why scale produced new capabilities, what emergence means and does not mean, and where scaling has started to slow.
📄️ 2. The three training stages
Lesson 2 of the free LLM course: what each of the three training stages contributes, why RLHF changed usability, what a system prompt really is, and why jailbreaks keep working.
📄️ 3. Retrieval (RAG)
Lesson 3 of the free LLM course: what retrieval-augmented generation solves, how chunking and embeddings work, why retrieval quality decides everything, and when to fine-tune instead.
📄️ 4. Tools and agents
Lesson 4 of the free LLM course: how function calling works, what an agent loop is, why reliability decays with the number of steps, and which agent designs actually work in production.
📄️ 5. Choosing, costing, securing
Lesson 5 of the free LLM course: how to select a model, how token pricing really behaves, how to evaluate honestly, and why prompt injection has no complete fix.
📄️ 6. Recap and FAQ
Lesson 6 of the free LLM course: a synthesis of the five lessons, an architecture decision guide, twelve frequently asked questions, and a vocabulary you can use precisely.
📄️ 7. Quiz and attestation
Test your understanding of large language models with a free 5-question quiz, corrected and explained: next-token prediction, context windows, temperature, hallucination and how RAG changes what a model can answer.