Building, Deploying and Managing AI Agents at Scale
summary
A pragmatic engineering blueprint for building, deploying, and managing agentic AI on Google Cloud, focusing on scalable, secure, and cost-controlled operations. Covers serving with Vertex AI Agent Engine, session state and long-term memory using MemoryBank, orchestration with Google’s Agent-Development-Kit (ADK), integration approaches to external services and data, authentication/authorization challenges, governance and cost guardrails, and observability/evaluation to prove reliability and business value.
AI-written summary · about
chapters (16)
- 0:00AI agent platforms intro
- 1:55Scaling AI agents in production
- 3:43AI agent system principles
- 5:50AI agent architecture layers
- 6:55Runtime layer best practices
- 9:04Orchestration layer design
- 11:31Model layer choices
- 13:43Evaluation challenges
- 15:07Agent integration layer
- 16:51Observability requirements
- 19:38Infrastructure abstraction
- 21:02Platform maturity fit
- 22:47Platform necessity check
- 23:51Key takeaways
- 25:42Q&A: Accuracy
- 28:34Q&A: MVP mindset
related talks
22mHOW TO BUILD YOUR OWN AI AGENT: THE COMPLETE ARCHITECTURE FOR ENGINEERING TEAMS
45mUnifying AI experience for customers: Building multi-agent systems with Google ADK
17mAI TechOps: Data and Infra for AI Foundation to Scale and Manage 1000s of Cohesive Agents
45mCreate and deploy your first agent with Google Cloud Gemini Enterprise
22mAgentic AI for Engineering Leaders | Strategy & Execution Guide
44mBuilding Agentic AI Systems: From Transactions To Conversations
14m15 Months of building AI agents in 15 minutes
01h 34mHands-on AI workshop: Multimodal agents