At Amazon, I engineered and operated an ML-driven content-moderation pipeline for Alexa utterances, using SageMaker, Bedrock, and an internal inference service across three production regions.
I improved release safety with automated rollback and staged bake-time gates, and built observability with 49 alarms, 20 metrics, and an operations dashboard. I also diagnosed production incidents involving connection pools, false-positive alarms, and locale routing.
Previously, I provided production support at AOL- A Yahoo! Company and created an employee self-service portal at Ocwen financial using Java, Node.js, and React. I’ve also worked with SQL, AWS services, and CI/CD pipelines.
