Individual LLM agents, even with communication capabilities, require structured coordination to tackle complex, multi-step problems. This chapter addresses the challenge of orchestrating these agent teams. We will examine how to construct sophisticated workflows that define agent collaboration. You will learn about various orchestration models, including state-driven and graph-based approaches, methods for adaptive task planning, and strategies for ensuring system reliability and fault tolerance. Additionally, we will discuss the integration of human-in-the-loop processes and techniques for managing larger agent ensembles.