Foundations of multi-agent systems with ADK

Опубликовано 1 октября 2025, 23:50

Discover the fundamentals of Multi-Agent Systems and how Google's Agent Development Kit (ADK) simplifies their creation. This video introduces the core concepts of multi-agent systems, including decentralized control, local views, and emergent behavior. We then explore how ADK supports development through its distinct agent types: LLM, Workflow (Sequential, Parallel, Loop), and Custom agents. Finally, understand the foundational concept of agent hierarchy for structuring your multi-agent applications.

Chapters:

0:11 - Introduction to Multi-Agent Systems and ADK
0:57 - What are Multi-Agent Systems?
2:04 - How ADK Supports Multi-Agent Development
2:24 - LLM Agents
2:38 - Workflow Agents
3:13 - Custom Agents
3:35 - Agent Hierarchy
4:17 - Recap and Next Steps

Resources:
Google ADK tutorial → goo.gle/48GQ8VO
Google ADK doc → goo.gle/485dvIx
Google ADK sample → goo.gle/4pOnIPV


Watch more [series name] →
Subscribe to Google Cloud Tech → goo.gle/GoogleCloudTech

#GoogleCloud #AIAgents #ADK

Speakers: Annie Wang
Products Mentioned: Cloud General