LAB333: Evaluate Reasoning Models for Your Generative AI Solutions

Important

This repository contains training materials for a hands-on workshop on Reasoning models. It provides an overview of reasoning models (what they are, how they work, when to use them) and illustrates tradeoffs and best practices for usage with examples. You can complete this in self-guided mode at home - if you have questions or need help, post messages to this discussion thread - we'll monitor it regularly and respond within 24 hours.

Workshop Overview

Reasoning models excel at complex problem-solving tasks and nuanced analysis that benefit from multi-step analysis or processing - as opposed to simple question-answering. So, how do you know if a reasoning model is the right fit for your scenario? In this workshop, we'll take you from core concepts to code-based exploration of reasoning models on Azure AI Foundry.

• Duration: 75-mins
• Level: Beginner/Intermediate

Learning Objectives

This workshop is designed to jumpstart your journey into understanding, and working with, reasoning models in generative AI. By the end of the journey you should be able to:

1. Explain what reasoning models are, and how they work
2. Deploy and explore reasoning models using Azure AI Foundry and GitHub
3. Build intution for usage with low-code and code-first approaches
4. Gain insights into patterns and practices for practical applications

Quickstart

1. Fork the repo to your personal GitHub account to get your own copy
2. Launch GitHub Codespaces and work with a VS Code editor in the browser
3. Open a VS Code terminal when ready, run mkdocs serve > /dev/null 2>&1 &
4. Work through the instructions in GitHub Codespaces.

Questions or Comments?

We welcome feedback to help us improve the learning experience.

1. File an issue. We welcome feedback on ways to improve the workshop for future learners.
2. Join the Azure AI Foundry Discord. Meet Azure AI community members and share insights.
3. Visit the Azure AI Foundry Developer Forum. Get the latest updates on Azure AI Foundry.

Session Resources

Resources	Links	Description
Build session page	https://build.microsoft.com/sessions/LAB333	Event session page with downloadable recording, slides, resources, and speaker bio
Microsoft Learn	https://aka.ms/build25/plan/ai_agents_apps__data	Official Collection or Plan with skilling resources to learn at your own pace