Summary
Paul Duncanson is a Lead Data Scientist with 12 years of experience building production-grade ETL and ML pipelines across healthcare, finance, and cloud platforms. He has led architecture and engineering efforts at McKesson and Microsoft Research, delivering containerized, Spark/GPU-enabled frameworks that integrate GPT-class models and streamline model lifecycle from local to distributed deployments. His background spans applied NLP, fraud detection, IoT analytics, and real-time streaming (Kinesis/Kafka), with hands-on implementation in Scala, Python, and cloud-native tooling. Notably, he has a track record of turning research-grade ML into operational systems—designing in-house ML orchestration at Microsoft and cancer biomarker measurement pipelines at McKesson. Based in the United States, Paul pairs deep engineering breadth with pragmatism, mentoring teams and accelerating solutions from prototype to scaled production.
12 years of coding experience
10 years of employment as a software developer
BS Information Systems Management, BS Information Systems Management at University of San Francisco