I build end-to-end geospatial data science pipelines — combining satellite imagery, spatial econometrics, and machine learning to forecast wildfires, predict crop loss, and monitor welfare in data-sparse regions across the Global South.
I'm an Associate Professor in the Department of Geography & Environment at The George Washington University, where I also serve as Director of Graduate Studies. My work sits at the intersection of remote sensing, machine learning, and geospatial modeling — with applications ranging from fire probability in California to crop-loss prediction, mapping urban deprivation, and monitoring household welfare in data-sparse environments across the Global South.
Alongside research, I lead or co-lead two Python packages that have become widely used in the geospatial community: GeoWombat, which simplifies large-scale remote sensing workflows, and xr_fresh, which automates time-series feature extraction from gridded data. I'm also the lead author of pyGIS.io, a freely available online textbook that has become a common reference for geospatial programming in both academic and professional settings.
I've served as a consultant to the World Bank, the U.S. Department of Treasury, and private geospatial firms — building end-to-end pipelines that combine satellite imagery, spatial econometrics, and machine learning. My work has been supported by USAID, Meta, and the National Science Foundation, and published across PNAS, PLOS ONE, Climatic Change, Ecological Economics, and Remote Sensing.
Second course in the GIS sequence — spatial analysis, geoprocessing workflows, and cartographic design. Lab-based with real policy and environmental datasets.
Python for geospatial analysts — scripting, automation, and data pipelines. Foundations of GeoPandas, rasterio, and xarray for reproducible spatial work.
Graduate GIS with Python — advanced spatial data structures, scripting, and reproducible analysis workflows. Final project is student-designed.
Advanced open-source geospatial programming. Large-scale remote sensing workflows, time-series feature extraction, and machine learning with GeoWombat and xr_fresh.
Over 15 years of consulting with public-sector agencies and private geospatial firms — building end-to-end pipelines that combine satellite imagery, spatial econometrics, and machine learning. Engagements typically span scoping, data architecture, model development, and delivery of reproducible workflows to client teams.
Recent work has covered crop-loss forecasting, urban deprivation mapping, electricity reliability from night-lights, and agricultural productivity in data-sparse regions.
Request for service: Please fill out this form
