Swiss Data Science Days 2025
Swiss Data Science Days 2025
Author: Ruwen Frick
On June 26, 2025 I had the pleasure of attending the IEEE Swiss Conference on Data Science at the beautiful Circle Convention Center at Zurich Airport. On the agenda for the day stood two very promising workshops: “From Queries to Actions: Hands-on With RAG and AI Agents” and “From Garbage to Gold: Building Trust in AI Through Data Quality”.
Morning
From Queries to Actions: Hands-on With RAG and AI Agents
Abstract
Ready to elevate your AI applications beyond document retrieval? Join this hands-on workshop to build intelligent systems that combine the precision of Retrieval-Augmented Generation (RAG) with the autonomy of task-oriented AI agents.
Start by implementing a RAG pipeline for accurate information retrieval, then advance to creating agents that autonomously plan and execute complex tasks. Through guided exercises and real-world examples, you will learn when to use each approach and how to combine them effectively.
By the end of the session, you will have built a hybrid system capable of precise data retrieval and autonomous task execution-ready to tackle real-world challenges.
Organizers
Elena Nazarenko, Lucerne University of Applied Sciences and Arts (HSLU) Aygul Zagidullina, Lucerne University of Applied Sciences and Arts (HSLU)
My resume
My expectations for this workshop were rather high, as I hoped to learn new advanceded RAG techniques I could then apply to the projects I was currently working on. Building RAG Systems that leverage AI-Agents to extract structured output from heteregounes input data currently consumes a considerable amount of my workdays. The presenters from HSLU were lovely, but the level of the workshop was unfortunately a bit too low for me to really profit from it. Most of the things that were presented I was already familiar with.
Afternoon
From Garbage to Gold: Building Trust in AI Through Data Qualit
Abstract
Unlock the full potential of your AI initiatives by mastering data quality management. In this hands-on workshop, you will dive into real-world case studies to understand how data quality shapes AI performance and business outcomes.
Through hands-on exercises using industry-leading data quality testing frameworks, such as Deequ or Great Expectations, you will gain practical experience in implementing automated quality checks and solving real data challenges.
Collaborate with peers to share insights, tackle implementation hurdles, and leave equipped with actionable strategies to ensure your data supports trustworthy, high-performing AI. Perfect for anyone aiming to turn data into a strategic advantage.
Organizers
Fiona Hefti, Innovation Process Technology (ipt) Kilian Dresse, Innovation Process Technology (ipt)
My resume
This workshop showcased how the Great Expectations framework can be used effectively to establish and monitor data quality. Since poor data quality has often been the primary bottleneck in the machine learning projects I’ve worked on, I was excited to explore the capabilities that Great Expectations offers. The Jupyter notebooks provided were well-structured and gave us a valuable hands-on experience with real-world data. I’m looking forward to applying Great Expectations in future projects.