AI Agents in Action: BuildingSmart, OpenSource LLM Workflows

AI-powered autonomous agents are changing the way we interact with large language models (LLMs). In this hands-on workshop, you will learn how to build and customize AI agents using open-source models and frameworks. We will explore how these agents reason, plan, and execute tasks, and how they can be integrated with external tools or run using local models (e.g., via LM Studio or ollama).

This workshop is designed for developers, researchers, and AI enthusiasts who want to build intelligent, privacy-friendly LLM applications. No prior experience with AI agents is required—just basic Python knowledge and a curiosity to experiment!

--- What You’ll Learn ---
✅ What AI agents are and how they interact with LLMs
✅ How to build AI-driven workflows using open-source frameworks
✅ How to run agents on local language models for privacy-friendly solutions
✅ How to integrate external tools and APIs into agent reasoning
✅ Bonus: If there’s interest, we can touch on synthetic data generation as a tool for testing and evaluation

AI agents allow large language models to go beyond simple text generation—they can think, plan, and take action. In this workshop, we will explore how to design LLM-powered agents that automate workflows, process information, and interact with external systems.

The focus will be on open-source AI and giving you hands-on experience with local and cloud-based models. You’ll start by building a simple AI agent, then extend it with new capabilities, such as integrating APIs or working with structured data.

--- Key Topics Covered ---
✔ Understanding AI Agents: What they are, how they work, and common frameworks
✔ Hands-on Development: Building a simple agent from scratch
✔ Expanding Capabilities: Enhancing agents with plugins, APIs, or external tools
✔ Running Local LLMs: Deploying AI agents using models on your own device
✔ Optimizing Agent Performance: Debugging, logging, and prompt tuning
✔ (Optional Discussion): If participants are interested, we can explore synthetic data generation as a tool for evaluating and testing AI workflows

--- Who Should Attend? ---
👨‍💻 Developers interested in LLM-powered automation
📊 Data scientists & AI researchers exploring autonomous AI workflows
🛠️ Tech enthusiasts looking for hands-on experience with AI agent frameworks

--- Prerequisites ---
• Basic Python programming skills
• Familiarity with Git & working in a terminal
• Some knowledge of LLMs (not required but helpful)
• A laptop with Python 3.11+ installed, or the ability to use a cloud-based environment

Luca Rolshoven

I am a PhD student doing research in NLP. I am working on an interesting Sinergia project that aims to measure the level of sustainability in call for tender documents in the domain of public procurement. Previously I worked as a Senior Machine Learning Engineer at BSI, where I was focusing on the deployment and ethical use of Large Language Models within the BSI Customer Suite. Regarding my studies, I earned my masters's in Computer Science, specializing in Data Science, from the University of Bern. During my master's program, I contributed to the development of DiBB, a Python library aimed at distributed black-box optimization.

Outside of work, I keep a close eye on emerging trends in generative AI and human-centered technology. When I'm not exploring the digital realm, you'll find me delving into psychology, decoding escape rooms, and indulging in bicycle rides.