AI Fine-Tuning in Microsoft Foundry to Make Your Agents Unstoppable

Explore advanced AI fine-tuning techniques in this 45-minute conference session from Microsoft Ignite 2025 that demonstrates how to build production-ready AI agents using Microsoft Foundry. Learn practical steps for choosing models, generating training data, and evaluating performance while discovering the latest synthetic data generation features that automatically create training datasets. Master multi-agent collaboration approaches for high-quality data creation and understand when fine-tuning makes the most impact for tool calling, data extraction, and workflow execution. Follow along with real-world demonstrations including launching RFT in Foundry using GPT-5 and O4 Mini, building AI-powered contract repositories, and creating scalable AI systems capable of processing millions of documents daily. Gain insights into complex query handling, AI extraction techniques for deriving key contract details and relationships, and how to transform models into agents that deliver accurate, consistent results in production environments. The session covers both Azure OpenAI and open-source model customization within Microsoft Foundry, providing comprehensive guidance for developers looking to build agents that perform rather than just respond.