Plan before you build: Deterministic planning patterns for AI agents

Tired of agentic AI that fails or wastes API budgets? Building reliable, production-ready agents requires deterministic planning, not probabilistic guesswork. This session explores the full spectrum of planning patterns—workflows, HTN, GOAP, utility AI, and supervisor—and how to combine them into robust, hybrid architectures. Learn how to maintain control over agent behaviour by making the plan the source of truth.