AI
AI Agents: Meet the Azure Cosmos DB vNext Emulator
Over a typical 2 day window, we record 10 times as many emulator sessions lasting more than 30 minutes for vNext as for the legacy emulator. Our bet is that agentic development is helping drive the difference! The appeal is straightforward: coding agents need a database they can run anywhere. They also need a database they can start, seed, test against and debug as they work. Our vNext emulator fits that workflow naturally. A different kind of developer With the vNext emulator, we're building for a broader idea of the developer: a person at a keyboard, a coding agent, or the two working together. The...
Azure MCP Server 3.0 stable release: More tools. More possibilities
Explore the new Azure MCP Server 3.0 stable release, with AI agent tools for Terraform, backup, recovery, and troubleshooting, plus stronger security defaults.
Introducing the Agent Experience (AX) Practitioner Playbook
AI coding agents now pick the SDK, the version, and the pattern for your developers. The AX Practitioner Playbook is our method for finding out whether they get your technology right, and for fixing the sources they rely on when they don't.
Dynamic Workflows in Azure Functions Hosted Skills: Durable, AI-Led Work for Event-Driven Apps
Azure Functions is the event-driven application platform, and Hosted Skills adds AI reasoning to your app. Dynamic Workflows give the app a durable way to carry out AI-led, multi-step work after the triggering invocation is finished. The model writes a plan (a DAG) one time, and Durable Functions runs it. In our benchmark, token use decreased by 56-93% and end-to-end latency decreased by 77-95%, with the same output quality.
AI Development on Windows: from PyTorch and llama.cpp to Windows ML
Today we are highlighting improvements across a few key open-source projects many of you already use, as well as updates to Windows ML. To empower developers, we must support the broad range of tools for experimentation and exploration across inference and training, in addition to our production grade native inference stack. Available today, Windows adds experimental llama.cpp support to Windows ML so you can run GGUF models locally through new task-specific APIs. Windows ML is the unified, high-performance local AI inferencing framework for Windows. Our experimental Windows-native Runtime API is now in preview ...
From Spec-First to Enterprise-Ready: Extending GitHub Spec Kit
How presets, extensions, bundles, workflows, and governed catalogs make Spec-Driven Development adaptable at scale This article continues Spec-Driven Development: A Spec-First Approach to AI-Native Engineering, which introduced the principles and lifecycle of SDD. Here, we focus on the next challenge: adapting that workflow to the standards, domains, tools, and governance needs of a large engineering organization. When a good workflow meets organizational reality Spec-Driven Development (SDD) gives teams a durable source of truth: structured specs that carry intent through requirements, design, implementatio...
Upcoming SQL and AI Events to Put on Your Calendar
Developers, database administrators, data professionals, architects and technology decision makers have several opportunities to connect, learn and gain hands-on experience this fall. 2027 Q4 SQL Events Get hands-on with SQL and AI The Friday workshops are led by the SQL product group! They provide a full day of practical learning before the community conferences begin. This AI workshop is designed for SQL developers, database administrators, application developers, architects and other data professionals who want to build AI-powered applications without abandoning the SQL tools and security model they alr...
Bring agentic reasoning to Python function apps with Azure Functions Agent bindings (preview)
Add bounded AI reasoning to Python function apps with Azure Functions Agent bindings, while keeping deterministic application code in control.
What is Agent Experience (AX)?
Your agent says it's done, and the code compiles, but did it pick your technology? Did it use it correctly? Learn what Agent Experience (AX) is, how to measure it, and why the obvious fix isn't always the right one.
From an Empty Folder to an AI Agent: Meet create-cosmos-agent
The first few minutes of a new AI project often feel productive. Create a chat endpoint, connect a model, send a prompt, and display the answer. Then the real questions arrive. Where should the agent store memory? How do we keep one customer's data away from another? How do we trace the source of an answer? What happens when an action requires approval? How do we test the application locally before creating cloud resources? And how do we move to Azure without rebuilding the project? Those questions inspired me to build create-cosmos-agent, an open-source CLI for creating production-oriented TypeScript AI...
