From the course: Hands-On AI: Building Agents with the Google Agent Development Toolkit (ADK)
Unlock this course with a free trial
Join today to access over 25,000 courses taught by industry experts.
Putting it all together: Advanced multi-agent orchestration with a coordinator
From the course: Hands-On AI: Building Agents with the Google Agent Development Toolkit (ADK)
Putting it all together: Advanced multi-agent orchestration with a coordinator
- [Instructor] We've journeyed through the building blocks of multi-agent systems in ADK, understanding how agents can follow sequences, execute tasks in parallel, and even perform iterative loops. We've also just explored how agents maintain conversational context using session, state, and memory. Now, let's witness the grand symphony. How does WanderWise take a complex user request, like, "I'm off to Milan this weekend, love fashion and food. What should I pack? Any cool events? And what's the weather like?" and turn it into a coherent, helpful response? This requires a conductor, a master planner. At the heart of our sophisticated WanderWise system is the wanderwise_coordinator_agent. Notice it's defined as a standard ADK agent, an LLM agent in this case, but its role is unique: to orchestrate. The coordinator agent doesn't do the specific tasks, like fetching weather or planning itineraries itself. Instead, its power lies in LLM-driven delegation. It's equipped with a set of…
Contents
-
-
-
-
-
-
(Locked)
Hierarchical designs: Delegation, routing, and orchestration7m 42s
-
(Locked)
Parallel agents: Concurrent execution for enhanced efficiency5m 7s
-
(Locked)
Loop agents: Iterative processes and refinement6m 27s
-
(Locked)
Conversational blueprint: Understanding session, state, and memory6m 25s
-
(Locked)
Context-aware agents: Practical session and state management4m 28s
-
(Locked)
Putting it all together: Advanced multi-agent orchestration with a coordinator5m 6s
-
(Locked)
-
-
-