2026-03-21 –, Yuchengco Hall 4th Flr. Y409 (Workshop Room 2)
The convergence of Location Intelligence and Artificial Intelligence—known as GeoAI—is reshaping how we interpret the physical world. From using satellite imagery to predict crop yields to analyzing GPS traces for logistics, applying machine learning to spatial data is becoming essential for modern data scientists. This two-hour workshop offers a practical roadmap for building end-to-end Geospatial Machine Learning pipelines in Python.
We begin by covering core geospatial concepts, moving beyond standard tables to understand coordinate reference systems, spatial resolution, and the differences between vector and raster data. Participants will learn why treating coordinates as simple X/Y columns can break models and how to preserve spatial relationships.
The workshop then focuses on data preprocessing—the most demanding stage of any GeoAI workflow. Using GeoPandas, Rasterio, and Shapely, attendees will perform spatial joins, engineer proximity features, and prepare satellite imagery for model ingestion.
Next, we address model training and validation, emphasizing the challenges of spatial autocorrelation. Instead of random splits, participants will implement spatial cross-validation techniques, such as block CV, to avoid geographic leakage.
Finally, we explore deployment strategies, including serving spatial models via APIs and monitoring concept drift caused by seasonal or urban changes. Attendees will leave with both a conceptual framework and a practical code template for taking geospatial data from raw input to production-ready ML solutions.
Ian is a driven data solutions developer specializing in the integration of geospatial intelligence across diverse systems and domains. His career spans research and engineering roles that bridge science, sustainability, and digital transformation. Ian has contributed to ocean renewable energy initiatives, led geospatial software development, and architected cloud infrastructure for a geoscience AI startup. He currently serves as a machine learning engineer at a solar design software company, advancing its mission to power the world with sunshine through intelligent, data-driven solutions.