SQL Server 2025 opens a new chapter: support for vector databases - a key component powering Large Language Models (LLMs) and modern AI solutions. But what exactly are vectors? How does text, an image, or code become a row of numbers? What are embeddings, and why are they so critical for semantic processing?

In this session, I’ll explore:
- how LLMs "understand" data,
- what embeddings are and how to create them,
- how to store and search vector data in SQL Server 2025,
- and why a vector database is more than just a new data type.

This session will be rooted in practical scenarios and live demos - no unnecessary math, just a clear focus on bridging traditional data approaches with cutting-edge AI capabilities. If you want to start building solutions today that are ready for tomorrow, this session is for you.