Next-Gen AI Insights Native Vector Search in SQL Server 2025

Abstract: SQL Server 2025 extends your existing data platform with a native vector search, enabling AI-powered similarity queries directly within the database. New vector data types and built-in index structures allow you to efficiently store and query high-dimensional embeddings (e.g., from NLP or image models). Thanks to optimized Approximate Nearest Neighbor algorithms, you’ll achieve millisecond-level response times—even with millions of vectors.

This feature unlocks a variety of use cases: you can implement real-time personalized recommendations for e-commerce, semantic analytics on text or image data, or anomaly-based security scans without the need for an external search infrastructure. Seamless integration with Azure hybrid scenarios also lets you scale your vector searches both on-premises and in the cloud. All of this runs on your proven SQL platform, complete with role-based access control and Always Encrypted support for sensitive embeddings.

Speaker: Torsten Strauß

Speaker Bio: Torsten (MVP, MCSE, MCT), CEO of inside-sqlserver and Principal Microsoft SQL Consultant at Sarpedon Quality Lab, specializes in Azure DevOps as well as the performance optimization and development of SQL environments. He has delivered over 300 presentations on these subjects at various national and international conferences and user groups. Torsten has been honored as a Microsoft Most Valuable Professional (MVP) for Data Platform since 2018, and Sessionize has recognized him among their most active speakers for three consecutive years.

Linked In
https://www.linkedin.com/in/torsten-strauss/

Blogging
https://inside-sqlserver.com/

Agenda:

• 12:15-12:30pm: Sign-in & network
• 12:30-12:35 pm: Announcements & Intro
• 12:35-1:35pm: Presentation
• 1:35-2:00pm: Q&A, wrap-up

Location:
Online. You must register for the link