An evening on controlling AI agents that can call tools, investigate systems and take real actions, hosted at Snyk's Euston office.
Hammed Ajibade, Lead Software Engineer at Pulse, asks how you actually revoke an AI agent: splitting read, recommend and execute permissions, where to put the human approval points, scoped credentials you can withdraw, and keeping an evidence trail across agent workflows. Gerald Crescione, who leads the AI Security Engineers Community at Snyk, makes the case for the role itself, combining software, security and AI engineering, and how to find and train people into it.
Free, food and drink provided. Doors 18:00, talks 18:30, social from 19:30, ends 20:30.
Why go: this is the operational end of agent security rather than the policy end. Best for platform and security engineers who already have agents touching production systems.
