Storing and querying text embeddings in a database it might seem challenging, but with the right schema design, it's not only possible, it's powerful. Whether you’re building AI-powered search, semantic filtering, or recommendation features, embeddings, and thus vectors, are now a first-class data type. So how do you model them well inside a databa...

Announcing Vector Type and Functions in Azure SQL Managed Instance Public Preview We are thrilled to announce that Azure SQL Managed Instance now supports Vector type and functions in public preview.  Vector support and functions have been available in Azure SQL in public preview since December. Check out the detailed announcement Azure SQL Man...

OpenAI recently introduced a powerful new feature for developers: structured output using JSON Schema via the parameter. This makes it possible to request responses from a GPT-4o model that strictly match a given schema—no free-text, no guesswork. If you're working with Azure SQL, this is a game-changer. Combined with the stored procedure and SQ...

In today's data-driven world, delivering precise and contextually relevant search results is critical. SQL Server and Azure SQL Database now enable this through Hybrid Search—a technique that combines traditional full-text search with modern vector similarity search. This allows developers to build intelligent, AI-powered search experiences directl...

Vector databases like Qdrant and Milvus are specifically designed to efficiently store, manage, and retrieve embeddings. However, many applications already use relational databases like SQL Server or SQL Azure. In such cases, installing and managing another database can be challenging, especially since these vector databases may not offer all t...

I'm happy to share a new GitHub repository with developers: azure-sql-db-aspire. This collection of eight hands-on examples demonstrates how to integrate SQL Server and Azure SQL with .NET Aspire, making it easier to build modern, cloud-native applications. Why .NET Aspire for SQL Server? .NET Aspire is a new opinionated, cloud-ready stack for .N...

Security is a significant topic today, and the ability to access a service requiring authentication without using an API key, password, or secret is a common request from those concerned about the security of a solution, which includes all of us. In today's digital landscape, cybersecurity threats are increasingly sophisticated and frequent, mak...

In the previous article of this series, it was discussed how embeddings can be quickly created from data already in Azure SQL. This is a useful starting point, but since data in a database changes frequently, a common question arises: “How can the vectors be kept updated whenever there is a change to the content from which they have been generated?...

Embeddings and vectors are becoming common terms not only for engineers involved in AI-related activities but also for those using databases. Some common points of discussion that frequently arise among users familiar with vectors and embeddings include: Let’s tackle each one of these questions one by one starting from the very...

The RAG (Retrieval Augmented Generation) pattern, which is commonly discussed today, is based on the foundational idea that the retrieval part is done using vector search. This ensures that all the most relevant information available to answer the given question is returned and then fed to an LLM to generate the final answer. While vector search...