Observe and evaluate agents for quality and safety in Microsoft Foundry

In this Azure Friday episode, Scott Hanselman and Tulika Chaudharie show how to take an Aspire application from local development to Azure App Service for Linux. They demonstrate how to create and configure an Aspire project, deploy its frontend, backend, and dashboard as App Service resources, and use the hosted Aspire Dashboard to understand the running application. They also explore how the Aspire agent integration can help generate the required configuration and simplify the deployment experience. Chapters 00:00 - Introduction 01:45 - Creating an Aspire project 03:35 - Configuring AppHost for Azure App Service 09:15 - Exploring the deployed Azure resources 09:50 - Viewing resources in the Aspire Dashboard 11:05 - Running the sample application 12:00 - App Service configuration and service discovery 14:20 - Accessing the hosted Aspire Dashboard 16:00 - Aspire configuration and deployment guidance 17:45 - Aspire agent integration 19:35 - Wrap-up Recommended resources Quickstart - Deploy an Aspire app Azure App Service Connect Scott Hanselman | Twitter/X: @SHanselman Azure App Service | Twitter/X: @AzAppService Azure Friday | Twitter/X: @AzureFriday Azure | Twitter/X: @Azure

See how a single Azure Database for PostgreSQL Flexible Server scales read-heavy workloads using read replicas and virtual endpoints — the same pattern OpenAI relies on to run ChatGPT on Postgres. Scott and Paula show a live read workload saturating a primary, then offload it to a replica with a single hostname change, and finish with replicas across Europe staying within milliseconds of the primary. It's read scale-out without sharding or re-architecting your app. Chapters 00:30 - Introduction 01:55 - Why read replicas: read/write asymmetry 03:45 - Architecture: primary, replicas, and virtual endpoints 05:30 - Demo: the cluster and a read-only replica 10:28 - Baseline: saturating the primary 13:53 - The flip: offloading reads to a replica 16:48 - Wrap up Recommended resources Read replicas in Azure Database for PostgreSQL Azure Database for PostgreSQL Demo script Connect Scott Hanselman | Twitter/X: @SHanselman Paula Berenguel | LinkedIn: paulaberenguel Azure Friday | Twitter/X: @AzureFriday Azure | Twitter/X: @Azure

In this Azure Friday episode, Scott Hanselman and Sajee demonstrate the Azure Cosmos DB Agent Kit — a skill you install with one command that gives your coding agent 100+ Cosmos DB best-practice rules across data modeling, partitioning, query optimization, and SDK usage and much more. Using a multi-agent fitness coaching app as an example, they show how the kit caught a missing partition key filter that was leaking member data across tenants, recommended hierarchical partitioning for multi-tenant scale, and fixed a fan-out query—all before the code shipped to production. Chapters 00:00 - Introduction 00:33 - Meet Sajee & overview of the Cosmos DB Agent Kit 00:50 - The problem: partition key & query mistakes that cost money in production 02:32 - How the Agent Kit works: one install, 100+ rules across 12 categories 04:52 - Demo setup: fitness coaching multi-agent app with Cosmos DB 06:26 - Showing the data: missing partition key filter exposes other members' data 08:23 - Agent Kit findings: SQL injection, singleton pattern, fan-out queries 10:46 - Indexing best practices & query optimization recommendations 12:34 - Applying the fix: correct results and single-partition RU cost 13:13 - Wrap up & how to get started Recommended resources Azure Cosmos DB Agent Kit Agent Kit Repository Connect Scott Hanselman | Twitter/X: @SHanselman Sajeetharan | Twitter/X: @sajeetharan Azure Friday | Twitter/X: @AzureFriday Azure | Twitter/X: @Azure

Azure HorizonDB is a new Postgres service on Azure built for scale, availability, and performance across any workload. In this episode of Azure Friday, we look at how HorizonDB delivers predictable performance with built-in zone resilience. We also explore new in-database AI features like AI Model Management and AI Pipelines, and walk through the developer experience in VS Code to build, query, and manage efficiently using AI. Chapters 00:00 - Introduction 00:47 - HorizonDB Architecture Overview 02:47 - Performance Demo 03:37 - AI Model Management 04:19 - VS Code Postgres Tools Overview 06:44 - Debugging Query Plans with GitHub Copilot 09:16 - AI Pipelines 14:32 - Wrap Up Recommended resources Learn Docs Azure Product page Connect Scott Hanselman | Twitter/X: @SHanselman Microsoft for PostgreSQL | LinkedIn: Microsoft for PostgreSQL Azure Friday | Twitter/X: @AzureFriday

Foundry Observability empowers developers to take agents from prototype to production with end-to-end observability. We demonstrate how evaluations, tracing, monitoring, and optimization work together to identify issues, measure quality, and improve outcomes over time. Through a live demo spanning the Foundry portal and VS Code, we showcase a practical workflow for building more reliable, production-ready agents. Chapters 00:00 - Introduction 01:33 - Foundry portal demo 06:58 - VS Code demo 17:43 - Wrap up Recommended resources Observability in Generative AI - Microsoft Foundry Build 2026 Resources Connect Scott Hanselman | Twitter/X: @SHanselman Azure Friday | Twitter/X: @AzureFriday

The new Azure file share experience is now generally available! In this episode of Azure Friday, we explore how the new management model simplifies the way you create and manage file shares, helping streamline your day-to-day operations. We also walk through the updated portal experience and highlight some of the key benefits of the new file share experience. Chapters 00:00 - Introduction 03:38 - Demo on portal 10:20 - Benefits 12:54 - Closing & learn more Recommended resources Learn Docs Azure Product page Blog post Connect Scott Hanselman | Twitter/X: @SHanselman Azure Friday | Twitter/X: @AzureFriday Azure | Twitter/X: @Azure

Scott Hanselman talks with Omar Shorbaji from the Anyscale engineering team about how Anyscale on Azure scales Python AI workloads from a single notebook to thousands of CPUs and GPUs. Built on Ray, the most widely adopted AI compute engine, Anyscale gives you a unified runtime to build, train, and serve, running directly on Azure Kubernetes Service without the complexity of managing Kubernetes. See a live demo that fine-tunes a vision-language-action robotics policy, with the metrics you need to push GPU utilization higher. Chapters 00:00 - Introduction 00:52 - Ray and the Anyscale platform 03:11 - Start of demo: Workspaces 04:38 - Running a job and viewing utilization metrics 05:24 - Choosing the right scale 06:53 - Abstracting Kubernetes on AKS 08:53 - Wrap up and where to learn more Recommended resources Learn Docs Anyscale on Azure Connect Scott Hanselman | Twitter/X: @SHanselman Anyscale | Twitter/X: @anyscalecompute Azure Friday | Twitter/X: @AzureFriday Azure | Twitter/X: @Azure

Explore Azure MCP Server and Azure Skills working together to extend AI capabilities across Azure services. In this video, we walk through the developer experience in VS Code, showing how to configure, run, and interact with Azure Skills seamlessly. Chapters 00:00 - Introduction 00:50 - What's Azure MCP 01:58 - Install Azure MCP 05:43 - VS Code User Experience 08:49 - VS Code Demo 14:25 - Azure Skills 18:36 - Wrap Up / Outro Recommended resources Learn Docs Azure Product page Azure Skills Connect Scott Hanselman | Twitter/X: @SHanselman Azure SDK | Blog: Azure SDK Blog Azure SDK | Twitter/X: @AzureSDK Azure Friday | Twitter/X: @AzureFriday Azure | Twitter/X: @Azure

In this Azure Friday session, we walk through how to get started with Azure SRE Agent and demonstrate how agentic workflows can be applied to real operational scenarios—from help desk tickets to cost analysis and reporting and anything in between. The session starts with context engineering fundamentals, followed by a live demo that shows how memory, personalization, incident management and scheduled tasks come together to reduce operational toil. We wrap up by creating an SRE Agent from scratch and discussing consumption, limits, and best practices. Chapters 00:00 - Introduction 01:05 - Context engineering and getting started 02:04 - Start of demo 03:23 - Getting to know you and memory 06:21 - Handle a help desk ticket 10:05 - Scheduled task for cost management analysis and reporting 12:22 - Creating an SRE Agent 13:50 - Agentic unit consumption and limits 14:35 - Wrap up Recommended resources Integrated docs Learn Docs Azure Product Page Try it out now Lab for the demo GA Announcement Connect Scott Hanselman | Twitter/X: @SHanselman Shamir Abdul Aziz | LinkedIn: linkedin.com/in/meetshamir Azure SRE Agent | YouTube: youtube.com/@AzureSREAgent Azure Friday | Twitter/X: @AzureFriday Azure | Twitter/X: @Azure

In this episode, learn how to migrate on-premises Oracle Database workloads to Oracle AI Database@Azure, where Oracle database services run on Oracle Exadata infrastructure located inside Azure datacenters. Then see how, once your database is in place, you can modernize faster by connecting Oracle data to Microsoft Fabric for analytics and building AI experiences with Foundry, Copilot Studio—using familiar Azure tools. Chapters 00:00 - Introduction 00:58 - What is Oracle AI Database@Azure 05:33 - Azure Portal experience 11:30 - Microsoft integrations (Fabric, Foundry) 14:00 - Agentic experience 15:54 - Wrap up & close Recommended resources Learn Docs Azure Product Page Connect Scott Hanselman | Twitter/X: @SHanselman Oracle AI Database@Azure | LinkedIn: linkedin.com/groups/14707004 Azure Friday | Twitter/X: @AzureFriday Azure | Twitter/X: @Azure