Welcome to the Agentic AI Engineering course repository! This repository contains a collection of Jupyter notebooks designed to support the agentic AI course, focusing on techniques for enhancing AI models with agentic capabilities.
You can find all the course notebooks in the notebooks directory. These notebooks cover various aspects of building agentic systems with models from setting up web search, connecting agents, adding human feedback, building your evaluations and monitoring to creating MCP servers, providing both theoretical background and practical, hands-on examples.
- Run Locally: You can clone the repository and run the notebooks on your local machine. To do this, ensure you have a Python installation with the necessary dependencies. You might run into some problems with a few notebooks, as we built them with Colab in mind!
- Run on Google Colab: Each notebook includes a link at the top to open it directly in Google Colab, making it easy to run without local setup.
- Audience: Designed for students and professionals interested in AI and natural language processing.
- Topics Covered: The notebooks cover foundational and advanced concepts of agentic AI development, including:
- Multi-source data collection
- ReAct reasoning loops
- Tool orchestration
- Huyman-in-the-loop feedback
- Multi-modal generation
- MCP-based editing
- LangGraph workflows
- and more...
Clone the repository and explore the notebooks at your own pace. Whether running them locally or ideally in Colab, these notebooks will guide you step-by-step, enhancing your learning experience.
They are part of the Agentic AI Engineering course built by Towards AI and Paul Iusztin. They do not represent the full course, missing most of the educational content and the project we built during the course. Check it out to learn more about building production-ready agentic systems!