AI-Powered Search in SQL Server 2025

Proposed session for SQLBits 2026

TL; DR

SQL Server 2025 introduces native vector support, enabling AI-powered semantic search directly inside the database engine. But what does that mean for data professionals, how does it actually work, and why should you care? In this session, we’ll break down the fundamentals of vector search and show exactly how SQL Server stores, indexes, and queries embeddings to deliver intelligent search capabilities.

Session Details

SQL Server 2025 introduces native vector support, enabling AI-powered semantic search directly inside the database engine. But what does that mean for data professionals, how does it actually work, and why should you care?

In this session, we’ll break down the fundamentals of vector search and show exactly how SQL Server stores, indexes, and queries embeddings to deliver intelligent search capabilities.

We’ll cover:

What embeddings are and how the vector data type works in SQL Server
How to generate and store embeddings using an external LLM
Performing searches using vector_search() and vector_distance()
How vector indexes (aka the DiskANN algorithm) work under the hood

We'll then bring it all together to build an intelligent, AI-driven application.

This session is ideal for data professionals and developers who want a practical understanding of SQL Server’s vector capabilities, with real demos and real use cases.

3 things you'll get out of this session

Overview of embeddings and their generation What the vector datatype is How vector search works in SQL Server

Andrew Pruski's other proposed sessions for 2026

New features in SQL Server 2025 - 2026

Andrew Pruski's previous sessions

Disaster Recovery Strategies for SQL Server

"Hope is not a strategy" - Google SRE Team motto Being able to recover our data in the event of an outage is a (if not the most) critical aspect of managing SQL Server. But how do we plan for the worst? What technologies are available to us? And how do we implement a disaster recovery strategy when the worst has just occurred? Join Microsoft Data Platform MVP, Andrew Pruski, in this session to delve into SQL Server disaster recovery strategies.

Chaos Engineering for SQL Server

How can we test potential weaknesses in SQL Server? In this session we will explore the concept of Chaos Engineering, its core concepts, and the methodologies behind it.

Why is understanding Kubernetes important for your career as a Data Professional?

A panel discussion on why both - admins and developers - should understand the impacts and benefits of containerized applications.

Docker Deep Dive

Played around with SQL Server in containers and want to delve further into the platform? This session is for you! In this session we'll dive straight into the Docker platform and investigate its internals.

Container questions that Andrew and Anthony DM each other

A Q&A panel session with Andrew Pruski and Anthony Nocentino talking about all things containers! Hosted by Rob Sewell

Introduction to partitioning

Session will be aimed at Database Administrators\Developers who have not previously implemented partitioning within an OLTP database and is designed to give a overview into the concepts and implementation.

Containers and Clones: giant databases on tiny HDDs

Provision dev databases quickly and easily using containers and clones

An introduction to SQL Server in Containers

Containers are a new and exciting technology but what does it mean for DBAs? This session will give an introduction into SQL Server running in containers and what options are available.