AI Agents Don't Read Your Policy Docs. They Hit Your APIs — Gravitee
AI Engineer — Oct 10, 2026 — talk
Why watch
A technical deep-dive demonstrating why AI agent governance must live in API gateway infrastructure rather than in prompt instructions. Sam runs five live demos showing a gateway blocking destructive operations, preventing data exfiltration, rate-limiting buggy agents, caching repeated queries, and rejecting oversized prompts before they hit LLMs. Covers observability, vendor decoupling from hyperscalers, and tackling shadow AI. For platform engineers and developers building AI agent infrastructure.