Uploaded September 2026 | Updated November 2026, 3 hours ago
Part of the Modern AI Apps with Azure Cosmos DB Series.
From REST endpoints to serverless APIs, coding agents can dramatically speed up backend development. This session demonstrates how to use an agent to scaffold APIs, integrate Azure Cosmos DB SDKs, implement CRUD operations, and refine the architecture for real-world use.
Presenters:
Sajeetharan Sinnathurai - Azure Cosmos DB Principal Program Manager - linkedin.com/in/sajeetharan
Jay Gordon - Azure Cosmos DB Senior Program Manager linkedin.com/in/jaygordon0042
• Start with the Agent Kit — Give your coding agent the context it needs to build Cosmos DB apps the right way - https://aka.ms/CosmosDB-agentkit
• Subscribe to Azure Cosmos DB on YouTube - youtube.com/AzureCosmosDB
• Follow Azure Cosmos DB on X - twitter.com/AzureCosmosDB
• Follow Azure Cosmos DB on LinkedIn - linkedin.com/company/azure-cosmos-db
00:00:00 Introduction to the stream
00:03:04 Housekeeping and code of conduct
00:04:42 Defining the API contract and goals
00:06:33 Setting up the development workspace
00:10:14 Integrating the Azure Cosmos DB agent kit
00:15:12 Scaffolding the FastAPI project structure
00:23:42 Defining data models with Pydantic
00:29:00 Configuring best practices and partition keys
00:38:25 Implementing the first vertical slice (CRUD operations)
00:48:57 Conducting a production-minded code review
01:01:31 Testing the API with the Cosmos DB emulator
01:06:30 Closing remarks and look ahead to the next session
#azurecosmosdb #aicoding
[eventID:27410]
Part of the Modern AI Apps with Azure Cosmos DB Series.
From REST endpoints to serverless APIs, coding agents can dramatically speed up backend development. This session demonstrates how to use an agent to scaffold APIs, integrate Azure Cosmos DB SDKs, implement CRUD operations, and refine the architecture for real-world use.
Presenters:
Sajeetharan Sinnathurai - Azure Cosmos DB Principal Program Manager - linkedin.com/in/sajeetharan
Jay Gordon - Azure Cosmos DB Senior Program Manager linkedin.com/in/jaygordon0042
• Start with the Agent Kit — Give your coding agent the context it needs to build Cosmos DB apps the right way - https://aka.ms/CosmosDB-agentkit
• Subscribe to Azure Cosmos DB on YouTube - youtube.com/AzureCosmosDB
• Follow Azure Cosmos DB on X - twitter.com/AzureCosmosDB
• Follow Azure Cosmos DB on LinkedIn - linkedin.com/company/azure-cosmos-db
00:00:00 Introduction to the stream
00:03:04 Housekeeping and code of conduct
00:04:42 Defining the API contract and goals
00:06:33 Setting up the development workspace
00:10:14 Integrating the Azure Cosmos DB agent kit
00:15:12 Scaffolding the FastAPI project structure
00:23:42 Defining data models with Pydantic
00:29:00 Configuring best practices and partition keys
00:38:25 Implementing the first vertical slice (CRUD operations)
00:48:57 Conducting a production-minded code review
01:01:31 Testing the API with the Cosmos DB emulator
01:06:30 Closing remarks and look ahead to the next session
#azurecosmosdb #aicoding
[eventID:27410]
![Introduction to Copilot & Cowork Extensibility with WorkIQ Developer Tools
Explore the Copilot and Cowork extensibility landscape and discover where WorkIQ Developer Tools fit into the developer ecosystem.
Learn how developers can approach building plugins that connect enterprise knowledge, systems and actions to Copilot experiences, and get familiar with the tools and resources available to help you get started.
0:00 Introduction and Housekeeping
0:56 Welcome to WorkIQ Developer Tools Series
2:07 Introduction to Plugins and Components
2:45 Plugin Demo: Retirement Radar
5:12 Plugin Architecture: M365 vs. Anthropic
6:02 Understanding the Plugin Model
7:45 Plugin Registry Overview
11:45 Plugin Registry Demo
17:39 Introduction to WorkIQ Developer Tools (Wicket)
22:56 Wicket Demo: Building with GitHub Copilot
29:43 Validating and Improving Plugins
35:55 Publishing Plugins for Organizations
40:21 Q&A: Copilot, Coding, and Agents
48:08 The Future of the Code Tab
52:21 Advanced Topics and Closing Remarks
Get started: https://microsoft.github.io/wiqd
Practical guide: https://aka.ms/wiqd/cookbooks
[eventID:27604] Introduction to Copilot & Cowork Extensibility with WorkIQ Developer Tools](https://i.ytimg.com/vi/27cA73O6Fn8/mqdefault.jpg)
![O que criamos, o que funcionou e o que aprendemos
Raramente aprendemos mais com a parte de um projeto que funcionou perfeitamente.
Nesta sessão, pessoas da comunidade vão mostrar o que criaram, o que as surpreendeu, o que deu errado e o que fariam diferente da próxima vez.
Para mais informações sobre o Data Days e tudo o que ele tem a oferecer, acesse https://aka.ms/DataDays
📌 Este evento faz parte de uma série. Saiba mais aqui: https://aka.ms/AIatWork/ShowIt/y
#MIcrosoftReactor #LearnConnectBuild #Fabric #DataDays #AI
[eventID:27740] O que criamos, o que funcionou e o que aprendemos](https://i.ytimg.com/vi/2H2wBb1FWuk/mqdefault.jpg)
![Production Ready: Observability, Governance and Deployment
The last episode covers what you need before any of this runs somewhere real. OpenTelemetry traces and logs show exactly what the agent did on each turn. Purview handles governance and data protection. Evaluation gives us a way to tell whether quality holds as the SDK keeps moving.
Then we deploy to Foundry Hosted Agents so the thing runs as a service instead of a console app. We close by walking through the full refresh-prompt workflow as a maintenance pattern worth stealing.
📌 This event is part of a series. Learn more here: https://aka.ms/FMTA/series
🔗 Learn more by exploring the resource: https://aka.ms/AgentHarness
0:04 Introduction and Series Recap
3:12 MCP Tools and Agent Integration
9:26 Persisting and Serializing Agent Sessions
14:48 Running Local Models with Foundry Local
20:21 Observability
34:51 Governance and Data Policy
41:26 Evaluation Frameworks
45:56 Deployment to Foundry Hosted Agents
52:21 Closing Summary and Resources
#MicrosoftReactor #LearnConnectBuild
[eventID:27520] Production Ready: Observability, Governance and Deployment](https://i.ytimg.com/vi/2cG98BhvWDU/mqdefault.jpg)
![Construye tu forma de trabajar con Copilot: integra la IA en tu día a día
En esta sesión práctica aprenderemos a conectar diferentes formas de utilizar Microsoft Copilot y construiremos juntos un framework sencillo para diseñar una forma de trabajo potenciada por IA, desde una necesidad hasta un resultado concreto.
Trabajaremos con seis elementos: necesidad, prompt, proceso, asistente, decisión y resultado. A través de ejemplos y demostraciones en vivo, integraremos lo aprendido durante la serie para diseñar un flujo de trabajo en el que Copilot pueda ayudarnos a analizar información, ejecutar tareas, estructurar ideas y apoyar decisiones.
El objetivo no es utilizar IA en cada actividad, sino aprender una metodología que permita identificar dónde Copilot aporta valor, cómo integrarlo de manera consciente y cómo construir una forma de trabajo que podamos probar, mejorar y adaptar a nuestras propias necesidades.
🔗Aprende más accediendo a los recursos:
https://aka.ms/PromptAssitant
https://aka.ms/PromptGalleryLearn
📌 Este evento es parte de una serie, aprende más aquí: https://aka.ms/BuildWithCopilot/serie
#MicrosoftReactor #LearnConnectBuild #Copilot
[eventID:27612] Construye tu forma de trabajar con Copilot: integra la IA en tu día a día](https://i.ytimg.com/vi/2jcBFbQ8QSY/mqdefault.jpg)
![From Insight to Action: Build a Real Workplace App
Let’s start with a problem people actually have at work and build something useful around it. We’ll go from data and insight to a working Fabric App, showing the decisions, tradeoffs, and lessons along the way.
📌 This event is part of a series. Learn more here:
https://aka.ms/DataDays/FabConRecap/y
https://aka.ms/DataDays/IntoCerts/y
https://aka.ms/AIatWork/UnderstandIt/y
https://aka.ms/AIatWork/BuildIt/y
https://aka.ms/AIatWork/ShowIt/y
[eventID:27723] From Insight to Action: Build a Real Workplace App](https://i.ytimg.com/vi/2r-09tOixVg/mqdefault.jpg)
![AI Genius | エピソード3 レガシーシステムのモダナイゼーションから AI 運用まで
セッションの 紹 介
レガシー アプリケーションの なモダナイゼーションは 変 革 の 第 一 歩
AIシステムを 環 境 安 定 運 用 継 続 監 視 迅 速 障 害 対 応 高 信 頼 性 現 仕 組 不 可 欠
GitHub Copilotを 用 自 己 修 復 型 エージェントや 運 用 インテリジェンスを AIシステム 運 用 紹 介
AIを 運 用 負 荷 軽 減 信 頼 性 拡 張 性 えたAI 運 用 実 現 なアプローチを 学
学 習 内 容
- Copilot Modernize CLI のレガシーアプリケーションの 評 価 やアップグレードを
- Stacked Pull Requests 自 動 マイグレーションのレビューを 率
- Azure SRE Agent 診 断 検 証 修 復 策
- トークン 効 率 復 旧 時 間 受 け 入 れ 率 運 用 効 果 可 視 化 追 跡
- Azure SRE Agent
- Copilot App Modernization Agent
- Modernize CLI
- Stacked Pull Requests
- Human - in - the - Loop ガバナンスパターン
取 り 組 んでいるチーム
- 環 境 でAIシステムを 運 用 開 発 者 やプラットフォームエンジニア
- エージェント 支 援 障 害 対 応 運 用 改 善 検 討 しているSRE ・ DevOpsチーム
- 拡 張性 信 頼 性 AI 運 用 実 現 指 技 術 リーダー
講 演 者
Yurie
[eventID:27638] AI Genius | エピソード3 レガシーシステムのモダナイゼーションから AI 運用まで](https://i.ytimg.com/vi/3WFNubIzyOA/mqdefault.jpg)
![Afinal, o que é Fabric IQ?
Fabric IQ parece importante, mas o que ele realmente faz?
Vamos explicar como ele usa o contexto do negócio para ajudar a IA a entender seus dados, como se conecta ao restante do Fabric e como isso pode funcionar em um cenário real.
Para mais informações sobre o Data Days e tudo o que ele tem a oferecer, acesse https://aka.ms/DataDays
📌 Este evento faz parte de uma série. Saiba mais aqui: https://aka.ms/AIatWork/UnderstandIt/y
#MicrosoftReactor #LearnConnectBuild #Fabric #AI #DataDays
[eventID:27731] Afinal, o que é Fabric IQ?](https://i.ytimg.com/vi/3eQy8522XIc/mqdefault.jpg)
![AI-Powered Developer Productivity with GitHub Copilot and VS Code
AI-assisted development is transforming how engineers build and maintain software.
This session showcases practical workflows using GitHub Copilot and VS Code, covering code generation, debugging, testing, documentation, and productivity techniques you can apply in your day-to-day work.
Attendees will learn how to enhance their development process with AI while maintaining efficiency, quality, and responsible practices.
Key takeaways:
- Effective prompting for software development
- AI-assisted debugging and testing
- Developer productivity workflows
- Responsible AI usage practices
📌 This is part of a series, learn more here: https://aka.ms/ProdReadySystems/series
0:00 Session Introduction and Housekeeping
2:53 Introduction to Responsible AI in Engineering (Ram)
4:59 Assessing AI Capabilities vs. Necessary Evidence
6:48 Useful Engineering Workflows for AI
8:41 Practical Example: Query Generation Issues
11:05 Recommended Workflow for AI Tasks
13:35 Privacy, Data Handling, and Fine-Tuning
15:22 Human Oversight, Evidence, and Accountability
17:41 Ensuring Reliability and Evaluation
19:36 Transparency and Disclosure
21:54 Permissions, Recovery, and Conclusion
26:36 Introduction to Building Production-Grade Agentic AI (Anamaka)
31:38 Query Generation Workloads (NL2KQL and Graph)
33:47 Why Retrieval-Augmented Generation (RAG) over Fine-Tuning
36:00 Context Retrieval and Schema Validation
43:48 Building an Incident Investigation (DRI) Agent
46:13 Prompt-Based Agent Architecture
49:56 Cost and Context Window Challenges
54:55 Moving to Agentic Workflows for Reliability
58:56 Summary: Prompt-Based vs. Workflow Agents
1:03:04 Closing Remarks and Feedback
[eventID:27361] AI-Powered Developer Productivity with GitHub Copilot and VS Code](https://i.ytimg.com/vi/3j0y2ExcF5o/mqdefault.jpg)
![Azure Incident Retrospective: AOAI/Foundry-Failures in Sweden Central (ID:7QL5-Z50)
On 29 September 2026, the Azure platform experienced a service incident (Tracking ID: 7QL5-Z50) which caused AOAI / Foundry – intermittent failures in Sweden Central.
After publishing the Post Incident Review (PIR) we invited impacted customers and partners to join this Teams Live Event to be part of a discussion with our engineering leaders and experts – to learn more about the incident, how we responded, and what we learned.
Customers who participated in the live event also had an opportunity to get any questions about the incident answered, through a Live Q&A moderated by our engineering subject matter experts.
Stay informed during issues
– configure Service Health alerts for notifications and invites: https://www.aka.ms/ASH-docs
Learn from previous events
– Health history in Azure Service Health shows impact to your services: https://www.aka.ms/ASH-portal
Prepare for the unexpected
– learn more about steps you can take to prepare for cloud incidents: https://www.aka.ms/IncidentReadiness
[eventID:27782] Azure Incident Retrospective: AOAI/Foundry-Failures in Sweden Central (ID:7QL5-Z50)](https://i.ytimg.com/vi/3nveXKuAYdg/mqdefault.jpg)
![De “isso é legal” para “eu realmente poderia usar isso”
É fácil assistir a uma demo de IA e pensar: “Legal.” Transformar aquilo em algo realmente útil no trabalho é outra história.
Nesta sessão, vamos começar com um problema real e pensar juntos em como transformá-lo em um cenário de IA que valha a pena explorar.
Para mais informações sobre o Data Days e tudo o que ele tem a oferecer, acesse https://aka.ms/DataDays
📌 Este evento faz parte de uma série. Saiba mais aqui: https://aka.ms/AIatWork/UnderstandIt/y
#MicrosoftReactor #LearnConnectBuild #Fabric #AI #DataDays
[eventID:27730] De “isso é legal” para “eu realmente poderia usar isso”](https://i.ytimg.com/vi/4QAx_Lji-6Q/mqdefault.jpg)
![Bicep 102: Beyond the Basics, Scaling Your Infrastructure as Code with Bicep
One Bicep file works well for one team and one environment, but it gets hard to share, review, and keep consistent once several teams and environments depend on it.
This session picks up where Bicep 101 ends and shows how to scale a Bicep codebase. I am a product manager at Microsoft on Azure Bicep and owning Deployment Stacks. Bicep is Azures native infrastructure as code language and a stateless alternative to Terraform, so there is no state file to store, secure, or keep in sync.
In the demo I split a single Bicep file into modules, replace hand-written resources with Azure Verified Modules, and publish a shared module to a private registry so other teams use the same version. I add a parameters file for each environment and user-defined types, so a wrong value fails before anything is deployed, and set linter rules for the whole repository in bicepconfig.json. I then use GitHub Copilot to bring resources that were built by hand into the same code and add a GitHub Actions pipeline that runs Stacks
What-If before every deployment, which confirms that adopting those resources changes nothing in Azure. The result is Bicep code that many teams can share and deploy the same way. Everything runs live from VS Code, GitHub, and the Azure CLI. What attendees will learn:
- How to structure Bicep with modules, Azure Verified Modules, and a private registry that several teams can share.
- How to keep environments consistent with parameters files, user-defined types, and linter rules.
- How to bring hand-built resources into code with GitHub Copilot and check every deployment with Stacks What-If.
[eventID:27836] Bicep 102: Beyond the Basics, Scaling Your Infrastructure as Code with Bicep](https://i.ytimg.com/vi/5iht3COlGSY/mqdefault.jpg)