All things Azure
Developer focused how-tos, use cases and solutions on Microsoft Azure
Featured posts
Codex Azure OpenAI Integration: Fast & Secure Code Development
Introduction You can now enjoy the same Codex experience in CLI or VS Code with Azure OpenAI support. We've contributed the following five pull requests to mak...
How to develop AI Apps and Agents in Azure – A Visual Guide
As organizations explore new AI-powered experiences and automated workflows, there's a growing need to move beyond experiments and proofs-of-concept to producti...
How to subscribe for GitHub Copilot Business without GitHub Enterprise: A Step-by-Step Guide
Did you know that now you can subscribe to GitHub Copilot Business without GitHub Enterprise (GHE), which means you can get GitHub Copilot Business as a standal...
Latest posts
From Lab to Live: A Blueprint for a Voice-Powered AI Sales Coach
Until recently, building real-time voice AI for production was challenging. Developers faced hurdles such as managing audio streams and ensuring low-latency performance. That landscape is changing. The general availability of the Azure Voice Live API marks a turning point, providing a unified abstraction layer that simplifies the development of real-time voice and avatar experiences. This shift inspired me to build a reference implementation called the AI Sales Coach to demonstrate how these new capabilities can be applied to solve a practical business challenge: skill development. A Practical Appli...
Teaching the LLM Good Habits: How Azure MCP Uses Best-Practice Tools
Why best practices matter (and why many people skip them) Let’s be honest: when you’re in the zone shipping code or wiring up infrastructure, “best practices” can feel like that nagging voice telling you to eat your vegetables. We all know they’re good for us… but in the rush to get things working, they’re often the first thing to fall off the plate. The result? Configurations that work today but explode tomorrow. Pipelines that deploy - until they don’t. Costs that spiral out of control. And sometimes, security gaps big enough to drive a container ship through. That’s why the Azure team invests so heavi...
Build your own Microsoft Docs AI assistant with Azure Container Apps and Azure OpenAI
Learn how to deploy a self-hosted AI assistant that leverages Microsoft Learn content via the Model Context Protocol (MCP) and Azure OpenAI. It’s fast, secure, and ready for developer use in real-world apps. Prerequisite This guide assumes you already have an Azure OpenAI resource provisioned in your subscription, with a deployed model (e.g., gpt-4, gpt-4.1, or gpt-35-turbo). If not, follow the Azure OpenAI Quickstart before proceeding. The goal Imagine being able to ask Microsoft Learn, "How do I monitor AKS workloads?" or "What are best practices for Azure Bicep deployments?" and instantly get a precise, ...
Azure Container Options: Matching Services to Operational Responsibility
Customers are looking to modernize their application on cloud and ask us for guidance on which Azure container services to choose. Should you choose A or B container services or A+B container services? Some choose both (A + B) or use A or B approach. Instead of trying to force fit applications on a container service(s), in this blog post, we show you how to think modular and choose the container service(s) that best meets your application needs and operational responsibility. In other words, you choose the right tool for the right job. In fact, many organizations are gradually transitioning from an A or B min...
How We Use AI Agents for COBOL Migration and Mainframe Modernization
Legacy modernization is one of the most frequently raised challenges customers approach us with. They want to eliminate technical debt and run applications cloud-native. Obviously, there are different levels of complexity in legacy modernization, but the most challenging cases remain mainframe languages like COBOL and PL/1. COBOL still powers mission-critical systems in banking, insurance, and government. Modernizing away from the mainframe comes with increased complexity due to fewer experts, rising maintenance costs, and the loss of institutional knowledge — and the millions of lines of code certainly don’t ...
Build Biosensing AI-Native Apps on Azure with BCI, AI Foundry, and Agents Service
Co‑Author: JD Chibuk, Founder Blueberry X We obsessively track our steps, heart rate, and sleep - but what about our mental state? 🧠 Most of us navigate life blind to how stress, focus, and cognitive overload show up in our physiology. Tools to help capture cognitive indicators can help us change how we think, feel and act in high stress situations for the better. 🤖 Meet Your Cognitive Copilot Imagine an AI agent that quietly observes your indicators relating to cognitive activity, learns your patterns, and supports you in real-time. With breakthrough technologies like fNIRS (functional near-...
Automating Secure and Scalable AI Deployments on Azure with HashiCorp
While the promise of AI continues to generate momentum, many organizations face a familiar challenge: getting AI projects beyond the prototype phase. According to Gartner, only 30% of AI initiatives make it into production, and RAND reports that up to 80% fail to deliver expected outcomes. The problem isn’t model quality — it’s platform readiness. To deliver AI successfully, you need more than cloud infrastructure. You need a repeatable, secure, and governed platform built for modern, data-intensive workloads. The Shift: From Running AI in the Cloud to Building the Cloud for AI Deploying AI workloads in...
Announcing General Availability of GitHub Copilot for Azure—Now with Agent Mode
Unlocking a New DevOps Era GitHub Copilot for Azure was first introduced at Ignite 2024 as a public preview, empowering developers, IT operators, and DevOps practitioners to leverage natural language to provision, configure, and manage resources within their own Azure environments. Since then, many have used GitHub Copilot for Azure as an AI-powered developer assistant that transforms how they build and manage applications on Azure, directly from Visual Studio Code. Early adopters have shared overwhelming feedback on how it accelerates their Azure development. It flattens the learning curve and helps users st...
AI Genius Challenge: Flex your AI skills and beat the clock!
If you’ve been following along with the AI Genius Series, you already know we’ve been on a journey, one filled with learning, experimentation, and discovery. We’ve explored the power of Retrieval Augmented Generation (RAG), dived into Copilot deployment, and unlocked the potential of Azure’s AI ecosystem. But now, it’s time to take all that knowledge and put it into action. Now, after popular demand, I’m pleased to launch the AI Genius Challenge. A fast-paced, hands-on virtual hackathon designed to test your skills, stretch your thinking, and ignite your creativity from everything you have learned in the seri...