agents-course
This repository contains the Hugging Face Agents Course.
What is agents-course?
This repository contains the Hugging Face Agents Course.
Overview of agents-course
What it is
This repository contains the Hugging Face Agents Course, a free educational resource that teaches how to build AI agents. The course is authored by Ben Burtenshaw, Joffrey Thomas, Thomas Simonini, and Sergio Paniego. It is divided into 4 units plus bonus sections, covering from the basics of agents to a final assignment with a benchmark. The course is licensed under Apache-2.0 and is open source.
Key features
- Course divided into 4 units: Unit 0 (Welcome), Unit 1 (Introduction to Agents), Unit 2 (Frameworks for AI Agents), Unit 3 (Use Case for Agentic RAG), and Unit 4 (Final Project).
- Bonus units: Fine-tuning an LLM for function-calling, Observability and Evaluation, and Agents in Games with Pokemon.
- Covers three frameworks: smolagents, LangGraph, and LlamaIndex.
- Includes a final project with automated evaluation and a leaderboard of student results.
- Prerequisites: basic knowledge of Python and LLMs.
- Open source with contribution guidelines and a BibTeX citation for publications.
Use cases
- Learning the fundamentals of AI agents and large language models.
- Building practical agents using lightweight (smolagents) or production-ready (LangGraph, LlamaIndex) frameworks.
- Fine-tuning an LLM to perform function-calling.
- Implementing Agentic RAG to enable agents to retrieve and respond with relevant data.
- Evaluating and tracing agent performance with observability tools.
- Exploring AI agents in game environments like Pokemon.
Getting started
See the repository README for prerequisites and how to access the course. No installation commands are provided.
License and pricing
Licensed under Apache-2.0. The course is free and open source.
Based on the project's public README and repository metadata.