Curious about building agentic AI systems, but you're not sure where to begin? In this deep dive session we'll start with the basics of a single AI service, including how to add RAG, guardrails, memory persistence, tools, MCP and observability. Once we've go that down, we'll go beyond a single AI service and see how implementing multiple AI services into agents opens a whole new world of possibilities. We'll cover a variety of agentic AI patterns such as agentic workflows, the supervisor pattern, and goal-based autonomy, human-in-the-loop, etc. And we'll wrap it up by combining these patterns into an advanced agentic AI system that we'll be able to deploy to Kubernetes.
This session will focus on Java and Quarkus LangChain4j, but the concepts are universal for anyone to follow.