what is ai-agents-tutorial fr?

This repository is a growing tutorial series about AI agents, maintained by Amit Shekhar, the founder of Outcome School. Rather than shipping runnable code, it works as a curated table of contents that links out to blog posts and videos covering the field step by step. The series starts with the basics that agents are built on: large language models, retrieval-augmented generation, the Model Context Protocol, fine-tuning, and quantization. From there it moves into what an AI agent actually is, how it differs from a plain chatbot, and the five core parts that make one up. Later entries dig into how function calling works inside a language model, the think-act-observe loop that lets an agent take multiple steps toward a goal, and named agent patterns such as the ReAct agent, the reflection agent, and the plan-and-execute agent, each explained with its own anatomy, a worked example, and common ways it fails. There are also entries on agent memory, on context engineering as distinct from prompt engineering, and on agentic retrieval-augmented generation, including a variant called GraphRAG that uses a knowledge graph alongside vector search. The series continues into multi-agent systems, covering how multiple agents can take on different roles and communicate with each other. The author says new blog posts and videos will keep being added over time, so the list of topics is meant to expand. There is no installation step because this is not a library, it is a linked curriculum meant to be read and watched in order, aimed at anyone trying to understand agents from the ground up rather than jump straight into a specific framework.