From LLM orchestration to autonomous agents: Agentic AI patterns with LangChain4j

In the last few months, LLMs have exploded from simple chatbots into active, autonomous components that can reason, plan, and act. Clement Escoffi shows you how to level up from basic prompt engineering to full-blown agentic AI using LangChain4j, a Java-native framework. You’ll learn key design patterns—memory, dynamic decision-making, tool usage, RAG, MCP and A2A integrations—plus how to string together multi-step goals for powerful AI workflows.

Through live coding with Quarkus, you’ll build and test an end-to-end agentic system from scratch, guided by real-world examples. Whether you’re aiming for task automation, intelligent assistants, or decision support, this session equips you with the architecture and practical tools to craft robust, maintainable autonomous agents in Java.

Watch on YouTube