https://alexvazquez.xyz Enterprise infrastructure security at Snap. Previously Security at Microsoft. Focused on securing AI in the corporate environment, plus data loss prevention, governance and zero trust. I build open-source tooling for the same problems. en https://alexvazquez.xyz/thoughts/policy-enforcement-over-sandboxing/ https://alexvazquez.xyz/thoughts/policy-enforcement-over-sandboxing/ Tue, 03 Nov 2026 00:00:00 GMT A sandbox cannot hold the work a company needs from an agent, so we enforce policy at the point of use instead. https://alexvazquez.xyz/thoughts/multi-agent-orchestration/ https://alexvazquez.xyz/thoughts/multi-agent-orchestration/ Mon, 03 Aug 2026 00:00:00 GMT What makes a fleet of coding agents beat a single one, and where the fleet quietly fails. https://alexvazquez.xyz/thoughts/ai-risk-and-dlp/ https://alexvazquez.xyz/thoughts/ai-risk-and-dlp/ Tue, 12 May 2026 00:00:00 GMT A coding agent acts with your credentials, so data loss prevention has to run on the machine where the agent runs.