At Tecton, I developed MLOps capabilities for multi-tenant machine-learning workloads, from model preparation through production deployment and platform operations. I also built highly available Python and Java infrastructure for shared ML services.
I processed telemetry and transaction data with Apache Spark, Hadoop, and AWS Athena to support analytics and ML feature workflows. I implemented Kafka and AWS Kinesis streams for live feature data, supporting inference paths reported at sub-50ms.
At Eccalon, LLC, I led delivery of ML platform services and data pipelines, coordinating priorities with applied scientists and engineering teams. I worked with data scientists on ingestion and scheduling improvements that contributed to a reported 45% increase in ingestion speeds.
I’ve also integrated OpenAI, Claude, LangChain, LangGraph, and RAG patterns into production Node.js services. Earlier, at United Health Group and Elastic, I developed healthcare data services and backend APIs, and supported testing and production reliability.
