This keynote outlines Google’s vision for the next era of AI, moving beyond using models with prompts to a vibrant, open ecosystem where AI agents that collaborate seamlessly. The talk details Google’s open-source contributions across three core pillars: 1) Open Models in Model Garden, offering developers choice for any task; 2) An Agent Development Kit to standardize and accelerate robust agent creation; and 3) A universal Agent-to-Agent (A2A) protocol enabling agents to communicate and interoperate freely. This session provides a roadmap for developers to build the next generation of interconnected, agentic AI on an open foundation.
Key Takeaways:
– Overall Stack for developing Agents
– How to leverage Open source Models in Agent Development
– Key Concepts of building with ADK
– How to create multiagent workflows using A2A