About

Mark Thornburg is an Applied Machine Learning Scientist at Amazon with over 7 years of experience in prototyping, training, and productionalizing machine learning models for large scale global marketplaces. His research areas include representation learning, recommender systems, pricing machine learning, and real-time fraud detection. Currently, Mark is the Applied Science lead for Amazon's competitive pricing deep learning service, which prices over 100 million ASINs across North America, Europe, and South/East Asia. He has also worked as an Applied Scientist at Uber, where he prototyped, trained, and productionalized a multi-target regression machine learning system powering competitive pricing for the US/CA/LATAM rideshare markets. The model outputs are core drivers of over 100 million dollars of promotional spend a year. Mark also worked as a Data Scientist, Applied Machine Learning at Postmates by Uber, where he designed, trained, and productionalized real-time fraud detection machine learning models for Postmates' delivery platform, serving US/LATAM markets. The models audit over 5 million deliveries a month for payment, invite, and courier fraud, flagging/cancelling high-risk deliveries before completion. The company was later acquired by Uber Technologies for 2.65B in November 2020. Mark holds a Bachelor's degree in Applied Mathematics and a Minor in Computer Science from the University of California, Berkeley. His tech stack includes Research Scientist, ML, Deep Learning, and Data Scientist. With 8.20 years of relevant experience, Mark is a highly skilled Applied Machine Learning Scientist with a proven track record of delivering successful projects in the industry.