When JSON Met Graph Part 2 of 4 - The Multimodal Database Series A note on naming. Throughout this series, when we say "SQL Server 2025" we also mean Azure SQL. The multimodal capabilities we discuss - native JSON, graph, vector, and columnstore - are available across both the on-premises engine and the Azure SQL family. In Part 1 we descr...
Part 1 of 4 - The Multimodal Database Series This is a four-part series about what happens when a single database engine handles relational, document, graph, vector, and analytical workloads natively - and what you stop paying for when it does. You spin up a database, point an agent at it, and start building. The first few tables go fast - user...
Applications often need to know what data looked like before. Who changed it, when it changed, and what the previous values were. Rebuilding that history in application code is tedious and error prone. This is especially valuable when exposing a database to an AI agent through MCP servers like SQL MCP Server, where information discovery matters. ...
Applications often need access to data without needing access to everything. Social Security numbers, email addresses, and phone numbers are common examples. Storing them is required. Exposing them broadly is not. This is especially valuable when exposing a database to an AI agent through MCP servers like SQL MCP Server, where safety and reversibil...
Sometimes applications need to remove data without actually losing it. Soft delete keeps rows in the database while making them invisible to normal application access. This is especially valuable when exposing a database to an AI agent through MCP servers like SQL MCP Server, where safety and reversibility matter. Learn more about SQL MCP Serve...
Supporting re‑ranking has been one of the most common requests lately. While not always essential, it can be a valuable addition to a solution when you want to improve the precision of your results. Unfortunately, there isn’t a universal, standardized API for a “re‑rank” call across providers, so the most reliable approach today is to issue a manua...
At the start of this year, we released a new set of T-SQL AI features for embedding your relational data for AI applications. Today, we have brought those features to Azure SQL and SQL database in Microsoft Fabric. This post will help you get started using the new AI functions of Azure SQL. Prerequisites Set up your env...
We’re excited to announce the General Availability (GA) of native Regex support in SQL Server 2025 and Azure SQL — a long-awaited capability that brings powerful pattern matching directly into T-SQL. This release marks a significant milestone in modernizing string operations and enabling advanced text processing scenarios natively within the ...
This article presents an AI-powered approach to automating SQL Server code analysis and refactoring. The system intelligently identifies inefficiencies and common T-SQL anti-patterns, applying best practices through a set of formalized coding rules, using prompt-driven instructions. It not only automatically rewrites problematic and inefficient...
I recently created a GitHub repository to share the T-SQL scripts I've accumulated over 10 years as a DBA. However, I've always struggled to find the right script when I need it. For years, I relied on simple shell commands like find or just my memory, which could take 10-30 minutes, if I was lucky enough to find it. But, we are in the AI age,...
