In the world of AI, running models continuously without usage is not just inefficient—it’s expensive. This session focuses on a powerful concept: AI systems should consume resources only when they are actively delivering value.
You’ll learn how to build scale-to-zero AI agents using Google ADK, Eventarc, and Cloud Run, enabling your applications to automatically scale down to zero during idle periods and scale up instantly when triggered by events.
We’ll explore how event-driven architecture combined with serverless infrastructure allows you to design intelligent systems that are both highly responsive and cost-efficient. This approach is especially valuable for modern mobile and backend systems where demand is unpredictable.
What You’ll Learn:
- What “scale-to-zero” means in AI systems and why it matters
- How to design event-driven AI architectures
- Using Google ADK to build structured AI agents
- Triggering workflows with Eventarc
- Deploying and scaling services using Cloud Run
- Cost optimization strategies for AI-powered applications
- Real-world use cases of serverless AI agents
Who Should Attend:
- Mobile & Backend Developers
- Cloud and AI Engineers
- Developers working with Google Cloud
- Anyone interested in building scalable and cost-efficient AI systems
By the end of this session, you’ll understand how to build intelligent, event-driven AI agents that scale dynamically—ensuring you only pay for what you use while delivering powerful user experiences.
