At TEKsystems, I architect real-time transcription and LLM summarization pipelines, using Node.js, TypeScript, AWS, Kafka, SQS, and PostgreSQL to improve latency, resilience, and documentation quality. I've reduced transcript latency by over 30%, cut diagnosis time by 45%, and helped increase transcription availability to 99.6% during provider outages.
Previously, I built backend services, event-driven analytics, ETL workflows, and observability systems for Opendoor and Slalom. I enjoy turning complex streaming, data, and automation requirements into scalable systems that improve product outcomes and developer productivity.
