Construire une IA agentique avec les Structured Outputs, Function Calling et MCP

Presentation by: Julien Dubois (Microsoft) 📕 Summary: AI agents are programs that act autonomously: to do so, they must be able to communicate programmatically with an AI and perform actions. In this session, we'll cover: Structured Outputs: How to force an AI to respond by following a JSON schema, so that this result can be mapped to Java objects Function Calling: How to define and call Java functions from an AI model MCP: The new protocol that standardizes how LLMs communicate with different data sources and tools We'll use the code, demos, and documentation I created to implement these features in LangChain4j using the brand-new Java SDK developed by OpenAI. Recorded in April 2025 at the Palais des Congrès, Porte Maillot, Paris. 🔥 To stay up-to-date on Devoxx France news, follow us on LinkedIn:   / devoxx-france   Bluesky: https://bsky.app/profile/devoxx.fr Visit our website: https://www.devoxx.fr/