This repository hosts the Microsoft Build 2025 lab content for Interacting with Multimodal Models and Agents in Azure AI Foundry
In this hands-on lab designed for developers who are getting started in their generative AI journey, attendees will get hands-on experience with Azure AI Foundry by interacting with the latest multimodal models from the model catalog. They’ll learn core concepts and prompt engineering techniques to effectively generate text, sound, and images, mitigating the risks of inaccurate or harmful responses. They’ll also create an LLM-powered agent prototype through UI, to combine generative AI capabilities with data connectors and powerful tools, enabling the agent to take actions on behalf of the user.
75 Minutes
- Understand how Large Language Models work
- Explore Prompt Engineering techniques and Retrieval Augmented Generation pattern
- Learn how to mitigate risks of harmful content in input/output to your model
- Interact with multimodal models through different interfaces
- Get started with building an Azure AI Agent
- Azure AI Foundry portal
| Resources | Links | Description |
|---|---|---|
| Build session page | https://build.microsoft.com/sessions/LAB324 | Event session page with downloadable recording, slides, resources, and speaker bio |
| Microsoft Learn | https://aka.ms/build25/plan/BestModelGenAISolution | Official Collection or Plan with skilling resources to learn at your own pace |
The instruction included in this repo are designed for users going through the lab in a provided Skillable environment, which is the MS Build lab attendees experience. For a self-guided learner experience please visit azure-ai-foundry/ai-tutorials.