Generalist Data Scientist in the Seattle area, with a specialty in clustering/segmentation and heuristic optimization. I'm interested in high impact business problems focused around capital-intensive processes.
I co-founded Cedar AI with 2 fantastic colleagues in summer 2017. We're tackling an unsolved problem in a mature and critically important industry in the freight-shipping world. We're bringing state of the art artificial intelligence into a live operational setting, which we predict will improve safety, lower cost, and decrease transit times.
15 internal papers and 2 patents
I was the team lead on the networking data science team at Amazon Web Services, where we used telemetry data produced by networks of routers and datacenters all over the world to forecast demand, ensure sufficient capacity, model/predict catastrophic failure, drive scaling decisions, and maximize operational efficiency.
Adding capacity in the cloud is a race at the right pace, allocating billions of capital just ahead of explosive demand while avoiding waste. We drove $2B+ in capital and built data-driven mechanisms to harness deployed capacity and improve utilization and unit cost.
12 papers and 350+ citations (Google Scholar profile)
My PhD focused on theoretical/quantitative biology, where I had the pleasure to explore paleontology, biogeography, epidemiology, complex networks, and the sociology of science. I completed a PhD from the University of Washington in 2013, where I worked with Carl Bergstrom. I developed methodology to delineate biomes from species distributional data in extant and fossil organisms. In general I'm fascinated by ecological processes that play out over millions of years and across vast swathes of area, such as the formation of rain forests, merging of seas, and blending of fauna.