Agent context engineering for production

Homegrown sessions and state management can get you started, and local development environments can rely on filesystems and Git. But for scalable, maintainable deployed agents, you need more. Every LLM turn is a compiled view across the current state, ongoing session context, memory, various artifacts, tool responses, examples, and instructions. Join this session to discover how to go from “it works” to “it works great at scale” with context engineering best practices, architectures, and services. Watch more: 100+ sessions from Google Cloud Next 26 → https://www.googlecloudevents.com/next-vegas/ Subscribe to Google Cloud Tech → https://goo.gle/GoogleCloudTech Speakers: George Lee, Kimberly Milam, Jeff Dixon, Preethi Prabhakar BRK2-098 #googlecloudnext