Built a Python humanitarian GIS analytics pipeline to turn fragmented conflict and displacement datasets into operational maps and timelines.
Stack: pandas, geopandas, shapely, fiona, pyproj, mapclassify, matplotlib, plotly, folium.
Architecture uses DuckDB for analytics GeoPackage for spatial layers, orchestrated via a Typer CLI and YAML run contexts across 15 country configs.
Curious how others structure Python GIS pipelines — DuckDB/GeoPackage locally, or straight to PostGIS?