Agentic Design Patterns

Agentic Design Patterns is an immersive, three-day workshop that explores the essential design patterns that underpin modern agentic systems — AI agents capable of reasoning, planning, and collaborating dynamically. Participants begin with foundational concepts, learning to construct robust agent architectures through the Pipeline, Routing, and Parallelization patterns. These day-one sessions focus on structuring agent workflows for efficiency, adaptability, and scalability through guided hands-on labs and practical implementation exercises.

Building on this base, the course advances into reflection, tool use, planning, and multi-agent collaboration patterns, enabling participants to create agents that plan intelligently, use external tools effectively, and coordinate with other agents to achieve complex objectives. The final day explores higher-order design strategies, including memory management, goal monitoring, exception recovery, and human-in-the-loop integration. Upon completion, participants will have both the conceptual framework and the applied skills to design and implement powerful, adaptable, pattern-driven AI agent systems.

This course is designed to provide a comprehensive overview of Agentic Design Patterns including:

• Design and implement agentic solutions using design patterns in Python-based agent frameworks
• Gain experience implementing a wide range of agentic patterns
• Select and apply appropriate patterns to real-world agent scenarios
• Understand multi-pattern strategies and use cases
• Work with LangChain, LangGraph, Crew AI, OpenAI, Google ADK and other agent frameworks