Claude on Google Cloud: Monitoring and Securing Agents at Scale

A co-hosted Anthropic and Google Cloud technical webinar on a problem every security and IT team now faces: autonomous agents can take actions across systems and make decisions without manual review, so how do you enforce what an agent is allowed to do, and how do you trace and audit what it actually did? The session works through both in a production context, building Claude agents with the Claude Agent SDK and wiring them into Google Cloud's security and observability stack.

Featuring Roy Arsan (Applied AI, Anthropic) and Ivan Nardini (Developer Relations Engineer, AI/ML, Google Cloud), attendees will learn how to give each agent a scoped identity and enforce per-agent and per-user policies with Agent Identity, Cloud IAM and Agent Gateway; how Model Armor detects and blocks prompt injection, tool poisoning and data leakage at runtime; and how the Agent SDK emits OpenTelemetry traces and audit events into Cloud Observability, with hooks and permission layers enforcing policy on every tool call. Live demos plus best practices. Recording made available to registrants after the event.