Geospatial Python developer with a hydrology background.
I turn raw geospatial data into clean, reproducible results — satellite imagery, climate grids, GIS layers, and messy tabular archives. I work end to end: from raw data to a correct, documented output file.
🛰️ Satellite & climate data — Sentinel-1/2, Landsat, ERA5-Land 🗺️ GIS automation — GDAL, rasterio, geopandas, QGIS, shapefiles & rasters 🌊 Hydrology & modeling — rainfall–runoff, discharge analysis, LISFLOOD 🤖 Applied ML — ConvLSTM, TCN for hydrological forecasting 📄 Data extraction — OCR pipelines for scanned documents and PDFs
LISFLOOD basin setup pipeline
End-to-end pipeline that turns a watershed polygon + DEM + soil data into a
complete, runnable LISFLOOD hydrological model, with multi-objective calibration
(Optuna). Config-driven, self-validating, no hard-coded paths.
python · lisflood · gdal · hydrology · geospatial
ConvLSTM rainfall–runoff
Deep-learning pipeline predicting river discharge from gridded daily meteorology
(ERA5-Land). Two-branch ConvLSTM + interpretability suite (saliency, Grad-CAM,
permutation importance).
python · tensorflow · machine-learning · remote-sensing
Freshet Detector
Detects snowmelt-flood timing, duration and volume from daily discharge series.
CLI + desktop GUI, baseflow separation, configurable detection modes.
python · hydrology · time-series
Python · rasterio · GDAL · geopandas · xarray · NumPy · pandas ·
TensorFlow · QGIS · NetCDF · Optuna