- Expert Track
- Online
- Upcoming
In this final workshop, you will dive into the world of AI agents and learn how they go beyond standard chatbots and simple LLM prompts to handle complex, multi-step tasks. You will explore the core architecture of modern agents—including tool integration, memory, and planning capabilities, and build a simple, notebook-based agent that can interpret goals and select the right actions autonomously.
We will examine real-world agent patterns, practical use cases, and current limitations, while highlighting key strategies for risk management and output verification. By the end of this session, you will understand how to design, test, and critically evaluate interactive AI agents for your own projects.
Please note: There will be a lunch break from 12:00 PM to 1:00 PM during this workshop.
Your Benefits
- Understand what AI agents are and how they differ from traditional chatbots and standard LLM-based applications.
- Explore the core components of an AI agent, including tools, memory, planning and multi-step workflows.
- Build and experiment with a simple notebook-based agent that can interpret user goals and select appropriate tools.
- Explore common agent patterns, practical use cases, limitations and risks, and learn why AI-generated outputs should be verified.
Instructor
The instructor is Kajol Raju, AI expert at the AI Service Centre (hessian.AI) specializing in providing AI computing infrastructure through application-oriented AI research.
Language
Workshops will be held in English.
Prerequisites
Basic knowledge of Python and familiarity with Jupyter notebooks. A general understanding of large language models is helpful.
Registration
You can register for this workshop and the other workshops of this Expert Track under the following link: https://redcap.kks.uni-marburg.de/surveys/?s=M8CC7REP3D333HC7.
Publications
TBA
