Session Abstract

When does it actually make sense to build an AI agent and when are you just adding complexity you do not need? This session focuses on practical use of AI agents with GitHub Copilot and Agent Mode, showing how to automate documentation, interact with data platforms such as Microsoft Fabric, and create reusable workflows. We’ll cover: -When to use agents instead of prompts or simple automations -Real examples for day-to-day work -Common mistakes (like overcomplicating with too many agents) -How to keep solutions simple and maintainable The goal: clear, practical criteria to help you use agents efficiently, no hype, just real data.

Samantha Cruz

I'm a Data Engineering Team Lead at isolutions Barcelona. I lead the design, delivery, and operation of end-to-end data platforms across Azure, Microsoft Fabric, and Databricks, with a strong focus on scalability, governance, and production reliability. I manage and mentor the data engineering team, define platform architecture standards, and work closely with stakeholders to translate complex business requirements into robust data solutions. In addition to hands-on technical leadership, I am an active speaker and educator, focused on sharing practical, experience-driven insights from real data platform implementations.