{"data":{"evidence":{"claim":"Fayed has built production AI-agent workflows that operate against real production infrastructure.","classification":"demonstrated professional experience","limitations":"This does not claim public chatbot-product ownership or foundation-model research.","related":["ml-infrastructure","developer-tooling","performance"],"slug":"agents","source":"FAYED_TECHNICAL_CONTEXT.md","source_freshness":"2026-08-21","strength":"Strong for production agent workflows tied to infrastructure operations.","support":["Agents inspect ML infrastructure state, validate cost attribution, and investigate infrastructure spend.","Agents help debug live failures in production ML ranking pipelines.","Alert-management agent workflows used human approval for consequential writes."],"title":"Production AI agents","when_supported":"Current role.","where_supported":"Meta production and ML infrastructure work."},"subject":"Fayed Nassor"},"description":"Fayed has built production AI-agent workflows that operate against real production infrastructure.","last_updated":"2026-08-21","title":"Production AI agents","url":"https://fayed.dev/evidence/agents"}