David Callender Email & Phone Number

David received a M.S. in Computer Science from Dartmouth College and Bachelor’s degree from Macalester College, where he double majored in Mathematics and Physics. His interests focus on Machine Learning (ML), algorithm development and optimization, heterogeneous computing, and hybrid database systems.David has developed Deep Learning algorithms for the identification of vasculature in medical imagery and for extracting weather and environmental information from mobile devices. In addition, he has applied Machine Learning techniques to acoustic and vibrational signal analysis, predicting GPS radio signal interference, analysis of geospatial data, and has employed Bayesian Active Learning to guide human medical tests.He has experience deploying ML models to both servers and mobile devices and developing RESTful APIs backed by secure heterogeneous structured and unstructured database servers. In addition, he has deployed and maintained database systems and servers supporting multi year medical studies spanning geographically distant and remote regions.He is well versed in algorithm optimization utilizing multi core CPUs as well as GPU and other hardware acceleration, including edge tensor processing units. He helped to develop an embedded high speed image processing application for deployment on an autonomous underwater vehicle and wrote the GPU accelerated C++ algorithm which combined custom fused CUDA kernels and optimized library routines to enable real time processing.Prior to joining Creare, David was ranked in the top half percent of the Kaggle Machine Learning community, which has since ballooned to over 3 million competitors. He placed in the top 1.4% of all teams in a global three-year, $3 million competition, applying machine learning to predict patient hospitalization using medical claim history. At Dartmouth College, he employed neural-probabilistic language models to learn medically relevant semantic relationships extracted from the Unified Medical Language System (UMLS) and to aid image classification. In addition, he studied the application of probabilistic graph models to discrete simulations to explore optimal strategies for the mitigation of influenza like epidemics in urban environments. Prior to Dartmouth, David worked at the University of Washington where he contributed to cutting edge research in the study of the folding of protein molecules. At Macalester he worked on research into the use of zinc-oxide thin films for photovoltaics and presented a thesis on the Mathematical foundations of Quantum Mechanics.

Listed skills include R, Machine Learning, Python, Java, and 15 others.