In this recorded Live! 360 session, you’ll learn how C# source generators work, why they matter, and how they can dramatically improve app performance by eliminating runtime reflection, reducing allocations, and enabling ahead-of-time (AOT) execution. Jim Wooley walks through real examples, from logging and regex optimization to JSON serialization, and shows how incremental generators improve build performance and keep generated code in sync as you type. You’ll also explore how to write your own source generator using Roslyn APIs, understand syntax trees and semantic models, and validate generators with unit tests. If you're building .NET apps and want to automate boilerplate, remove reflection, or tune your app for speed, this session gives you the foundations to get started. 🔑 What You’ll Learn • What source generators are and how they fit into meta-programming • How generators eliminate reflection and support AOT compilation • Real examples: logging, regex, JSON, validation, mappers • How incremental generators improve performance over older models • How to explore Roslyn syntax trees & semantic models • How to write your own generator using attributes & visitors • Unit testing and debugging approaches for source generators • Performance impacts and comparisons (AutoMapper, Mapster, mocking libs) ⏱️ Chapters 00:00 Intro + What source generators are & why they exist 02:54 Eliminating reflection and enabling AOT 05:44 How source generators differ from GenAI 08:05 Examples in .NET: logging, regex, JSON, validation 11:58 Demo: Logging generator & partial methods 14:46 Demo: Regex generator & performance benefits 16:14 Demo: JSON serialization generator 17:40 Live change-tracking: how generators respond to edits 19:20 Using attributes to trigger generation 20:00 Intro to Roslyn syntax trees & code analysis 26:05 Navigating syntax with the Syntax Visualizer 29:32 Understanding attributes & metadata via Roslyn APIs 31:00 Demo: Building a simple CSV generator 35:55 Classic (non-incremental) generator architecture 38:30 Walking the syntax tree & extracting semantic info 41:52 Generating new code with StringBuilder 43:10 Incremental generators & why they matter 45:30 Transformations, caching, and performance optimization 47:25 Using records and value semantics in generator pipelines 49:00 Executing generation & adding output to the compilation 50:55 Demo: Extended CSV generator with new properties 54:00 How to test generators with xUnit & Verify snapshots 58:30 Debugging generators (best practices & pitfalls) 1:01:00 How Roslyn builds trees, tokens, trivia & factories 1:01:49 When meta-programming makes sense (mapping, DTOs, reflection removal) 1:03:00 Performance comparisons (AutoMapper, Mapster, mocking frameworks) 1:06:20 Licensing, open source considerations & practical guidance 1:08:48 Q&A + extending the sample generator 1:10:03 Additional resources & where to go next 👤 Speaker: Jim Wooley Senior Principal, Slalom Consulting 🔗 Links • Jim's Roslyn Labs repo: https://msft.it/63325t2EwH • Explore more Live! 360 sessions: https://msft.it/63327t2EwJ • Check out upcoming VS Live! events: https://msft.it/63321t2Ew3 #csharp #dotnet #visualstudio #roslyn

In this recorded Live! 360 session, Eric Boyd reflects on 15 years of Microsoft Azure—from its earliest days to today’s global cloud platform—and explains what those lessons mean as organizations move deeper into cloud-native architectures and AI-driven development. Blending personal stories, real outage lessons, and hands-on customer experience, Eric walks through Microsoft’s Cloud Adoption Framework (CAF) and shows why strategy, skills readiness, governance, and DevOps discipline matter more than ever. He also looks ahead at what’s next across compute, data, and AI—including confidential computing, the growing infrastructure demands of AI, and why data readiness is critical for building effective AI solutions. 🔑 What You’ll Learn • How Azure has evolved over 15 years—and what early failures taught us • Why many cloud and AI projects fail without a clear strategy and framework • How the Cloud Adoption Framework (CAF) helps reduce risk and improve outcomes • Why DevOps, CI/CD, and Infrastructure as Code are essential in an AI-driven world • Azure security best practices: RBAC, managed identity, network isolation, encryption • How Azure’s global scale, regions, and availability zones support resilience • What’s next for cloud + AI: confidential computing, data platforms, and AI tooling ⏱️ Chapters 00:00 Introduction + celebrating Azure at 15 04:20 Early web experience + first distributed systems lessons 09:45 First look at Azure and the Professional Developer Conference (PDC) era 13:30 Early Azure outages and reliability lessons 17:58 Cloud Adoption Framework (CAF) overview 19:18 Strategy and planning: real customer examples 28:00 Ready phase: landing zones, naming, and governance 30:03 Adopt phase: DevOps, CI/CD, and Infrastructure as Code 36:03 Security fundamentals in modern Azure 41:00 Azure scale then vs now: regions, zones, and monitoring 45:25 AI’s impact on cloud architecture 50:24 What’s next: confidential computing and AI infrastructure 55:10 Preparing for the future: data, AI tools, and learning paths 👤 Speaker: Eric Boyd Founder & CEO, Responsive X | Microsoft Azure MVP 🔗 Links • Download Visual Studio 2026: http://visualstudio.com/download • Explore more Live! 360 sessions: https://aka.ms/L360Orlando25 • Join upcoming VS Live! events: https://aka.ms/VSLiveEvents #Azure #CloudComputing #MicrosoftAzure #AI

In this Live! 360 keynote, Brian Randell and Mickey Gousset unpack how agentic AI is reshaping DevOps—from coding and code review to automation, security, and incident response. You’ll see how GitHub Copilot, agent workflows, and new Azure capabilities let teams automate the repetitive work, keep humans in the loop where it matters, and accelerate delivery without sacrificing safety. Through real demos in GitHub, Azure DevOps, Visual Studio Code, and Azure’s new “agent” preview, you’ll learn how agents collaborate with developers and IT pros, where they excel, where they fail, and how to build a practical 30/60/90-day adoption plan for rolling this out in your own organization without losing control, quality, or security. 🔑 What You’ll Learn • What “agentic DevOps” means and how AI agents collaborate with humans • How GitHub Copilot Agents handle issues, PRs, refactoring, and multi-step workflows • Using Agent HQ to monitor, steer, and audit active agent sessions • How Copilot integrates with Azure DevOps for pull requests and automation • Security essentials: code scanning, secret scanning, and AI-powered autofixes • Responsible AI practices & the importance of keeping humans in the loop • How Visual Studio Code’s agent modes (Ask, Edit, Plan, Agent) improve developer workflows • Azure’s new autonomous agent for monitoring & incident response • A 30/60/90-day roadmap for bringing agentic workflows into your organization ⏱️ Chapters 00:00 Introductions & the evolution of DevOps 03:26 What “agentic DevOps” means 06:20 Demo: GitHub Copilot Agents in action 09:10 Agent HQ: monitoring and steering agents 12:30 Copilot + Azure DevOps workflows 18:00 AI models, trust, and human-in-the-loop practices 23:30 What’s new in VS Code for agentic development 30:20 Demo: reviewing AI-generated PRs 35:51 Security: code scanning, secret scanning & autofix 40:00 Real-world pitfalls & responsible AI 48:48 The productivity paradox & adoption challenges 51:32 30/60/90-day rollout plan 53:05 Azure’s autonomous agent for cloud monitoring 56:11 Final takeaways 🔗 Links • Explore more Live! 360 sessions: https://aka.ms/L360Orlando25 • Join us at upcoming VS Live! events: https://aka.ms/VSLiveEvents 👤 Speakers: Brian Randell & Mickey Gousset #devops #githubcopilot #agenticAI #azure #vscode

When a CI pipeline fails, developers often waste valuable time sifting through large test suites to identify the tests that need attention. In this session, I’ll introduce a tool I developed that automatically converts .trx test results into Visual Studio .playlist files directly within the CI pipeline. This allows developers to download the .playlist file from the CI pipeline and instantly start debugging locally, running only the tests that failed. A key part of developing this tool involved leveraging AI tools like GitHub Copilot and Microsoft Copilot to accelerate the process. These AI tools helped me quickly understand complex decompiled code, identify key differences between Visual Studio playlist versions, and iterate far faster than I could manually. I’ll demonstrate how AI can assist in the exploration phase, navigate lower-level details, and allow developers to focus on higher-value tasks, such as solving the core problem. By adopting similar AI-assisted workflows, you can speed up your own development, tackle complex problems more efficiently, and boost productivity in your projects. ➡️ Demo code: https://msft.it/63325tFTO7🎥 See more from .NET Conf: https://msft.it/63328tFTP6 🔗 Links: .NET 10 — The most intelligent and unified .NET yet: https://msft.it/63320tFTP8 .NET Aspire 13 — Simplified orchestration for cloud-native apps: https://msft.it/63323tFTP1 https://msft.it/63326tFTPy Core 10 — Modern, secure, and full-stack web development: https://msft.it/63328tFTPK .NET MAUI 10 — Build beautiful apps on every platform: https://msft.it/63321tFTP3 C# 14 — A more expressive and productive language: https://msft.it/63324tFTPu F# 10 — Safer, faster, and more functional than ever: https://msft.it/63326tFTPr Visual Studio 2026 — Faster, smarter, and built for the future: https://msft.it/63328tFTPp 🎙️ Featuring: Benjamin Michaelis (@BenMichaelis) Connect with .NET: Blog: https://msft.it/63321tFTPX Twitter: https://msft.it/63324tFTPw TikTok: https://msft.it/63327tFTPd Mastodon: https://msft.it/63320tFTPg LinkedIn: https://msft.it/63323tFTPc Facebook: https://msft.it/63325tFTPl Docs: https://msft.it/63327tFTPW Forums: https://msft.it/63329tFTPU 🙋‍♀️Q&A: https://msft.it/63320tFTPq 👨‍🎓Microsoft Learn: https://msft.it/63323tFTPt #dotnet #dotnetconf

See what developers learned at VS Live! Orlando and why 2026 is the year for immersive, hands-on Visual Studio training in Las Vegas or Redmond.

🌟Visual Studio 2026 is here 🌟 In this recorded Live! 360 keynote, Mads Kristensen and Nik Karpinsky walk us through The Road to Visual Studio 2026 — a faster, smarter IDE built for professional C#, C++, and .NET developers. You’ll see how the team re-architected Visual Studio for performance, reduced everyday “paper cuts,” and integrated GitHub Copilot and AI agents in ways that feel natural to your existing workflows—not disruptive. From faster startup times to AI-assisted debugging and modernizing older .NET apps, this session highlights how Visual Studio 2026 helps you deliver higher-quality code with less friction. ⌚Chapters: 00:59 — Visual Studio 2026 vision 06:16 — Removing the "paper-cuts" 07:30 — Smooth upgrade path 08:24 — Backwards-compatible extensions 09:20 — Modern Fluent UI with color themes 10:45 — Unified JSON-backed settings 12:00 — Quality of life changes 13:00 — Performance — faster and more responsive on the same hardware 19:33 — AI-assisted development with Copilot & agent tools 23:22 — Profiler agent diagnostics and performance optimizations 28:30 — Test & code review agents 30:30 — Modernization workflows & upgrading older .NET apps 34:44 — Decoupling the IDE and build tools 35:06 — Visual Studio Insiders, the newest features first 39:50 — Monthly updates & features 40:39 — Microsoft ecosystem updates 🔗 Links: • Download Visual Studio 2026: https://msft.it/63320tsVXe • Check out the release notes: https://msft.it/63324tsVXi • Explore more Live!360 sessions: https://msft.it/63326tsVXY • Missed Live!360 in Orlando? Join us at VSLive! in 2026: https://msft.it/63327tsVXl 👥Speakers: Mads Kristensen (@mkristensen) & Nik Karpinsky #visualstudio #visualstudio2026