Plan Before You Build: Deterministic Planning Patterns for AI Agents by Dan Dobrin @ Spring I/O 26

This talk at Spring I/O 2026 focuses on the importance of planning in the development of Agentic AI. The speaker discusses various planning patterns including workflows, Hierarchical Task Networks (HTN), Goal-Oriented Action Planning (GOAP), Utility AI, and Supervisors, emphasizing how these patterns can influence agent behavior and effectiveness. By exploring hybrid architectures that combine these approaches, developers can create more robust systems that maintain control while allowing for flexibility. Live demonstrations using Java 25 and Spring AI illustrate these concepts, highlighting the significance of a well-structured plan as the foundational element guiding agent interactions and compliance.