Alvin Ashcraft
Alvin is a technical writer and software developer specializing in Microsoft technologies, specifically .NET, C#, and XAML. He has over 30 years of experience developing applications for Windows, mobile, and the web. He has authored two books for Packt Publishing and is a full-time content developer on the Microsoft Foundry docs on Microsoft Learn. His published books are "Learn WinUI 3" and "Parallel Programming and Concurrency with C# 10 and .NET 6".
Proposals - Montreal 2027
Explore AI Dev Options with the AI Dev Gallery app
Do you want to learn how to add AI with local models and APIs to your Windows apps? In this session, we'll discuss AI scenarios and models including Phi, Mistral, Whisper, and more. The AI Dev Gallery is an open-source app in the Microsoft Store created to help Windows developers integrate AI capabilities in their own native apps.
A Dev's Guide to Microsoft Foundry: Models, Agents & SDKs
Unlock the full potential of Microsoft Foundry using your preferred language—whether it's .NET, Python, or JavaScript. This session dives into the code and tools required to build modern AI apps. We will explore three core pillars of development: - Models & Fine-Tuning: download, fine-tune & use models in your code - Building Agents: use Foundry agents for complex tasks - Use Azure AI Search & Image Analysis for context & vision
Build a local Windows AI model provider for GitHub Copilot
Learn how you can leverage the GitHub Copilot SDK, Windows App SDK, and .NET to build a model provider for GitHub Copilot in VS Code. Prompt your agents with the native Windows AI model built into Copilot+ PCs, leveraging your machine's NPU. All this in a packaged .NET console app project you can build and run with the winapp CLI in any editor or IDE.
Supercharge Your Agents: Custom Tools, MCP, and Foundry
Unlock your agents' full potential by connecting them to your data, APIs, and the web. This session is a practical guide to extending Microsoft Foundry agents using the latest standards. We will start with a tour of the powerful capabilities—including Code Interpreter and web search—before shifting focus to build your own. You'll learn how to leverage MCP to create standardized, interoperable tools that your agents can use to do real work.
