Geospatial Data Scientist

Role overview Develop geospatial and remote-sensing methods that turn imagery and spatial data into reliable analytical products within an existing intelligence platform. The role works primarily in Python across change detection, object detection, segmentation, and spatial-temporal analysis, spanning classical machine learning, statistical and image analysis, and deep learning approaches such as Vision Transformers and U-Nets.

Responsibilities - Develop change-detection workflows over open satellite time series, distinguishing meaningful change from seasonality, cloud and shadow effects, acquisition differences, and registration errors. - Build and evaluate object-detection, segmentation, and classification methods for satellite imagery, weighing modern deep learning approaches against simpler baselines in terms of accuracy, label requirements, generalization, inference cost, and production viability. - Assess suitability of different sensors, resolutions, acquisition conditions, and processing levels, extending methods across optical, multispectral, thermal, and SAR data as requirements develop. - Design preprocessing and feature extraction together with data engineers, including quality masking, compositing, co-registration, normalization, spectral indices, and sensor-specific corrections. - Combine raster outputs with vector and temporal data for spatial statistics, zonal analysis, anomaly detection, and comparison across areas and observation periods. - Build or source reference datasets and evaluation protocols, including spatially and temporally separated validation, false-positive and missed-detection measurement, and performance checks across regions and sensors. - Package tested Python methods for repeatable batch processing or inference and collaborate on runtime, memory, monitoring, and integration into platform workflows.

Requirements - Formal university training in geoinformatics, remote sensing, Earth observation, applied mathematics, computer science, or a related discipline, combined with substantial hands-on geospatial or remote-sensing experience. - Strong Python and scientific computing skills using tools such as NumPy, pandas, SciPy, xarray, GeoPandas, and Rasterio/GDAL. - Practical machine-learning experience with scikit-learn and PyTorch or an equivalent framework, plus TorchGeo or related geospatial deep learning packages. - Solid understanding of remote-sensing fundamentals including spatial, spectral, radiometric, and temporal resolution, coordinate systems, image alignment, and data-quality limitations. - Experience with satellite image analysis and at least one relevant task such as change detection, segmentation, object detection, or land-cover classification. - Sound statistical judgment around sampling, spatial autocorrelation, data leakage, class imbalance, uncertainty, and generalization; clear communication in English and eligibility to work in Germany.

Nice to have - Experience with thermal infrared imagery, SAR data processing and analysis, or combining observations from multiple sensors; a PhD in a relevant field. - Experience with geospatial foundation models such as Prithvi, TerraMind, or AlphaEarth Foundations embeddings and evaluating when they help over task-specific models. - Scaling geospatial analysis with Dask, Apache Spark/Sedona, or Zarr; deploying and optimizing GPU inference with PyTorch/CUDA, ONNX Runtime with TensorRT, or NVIDIA Triton. - Spatial SQL with DuckDB or PostGIS, STAC-based data discovery, and exploratory work in QGIS or geemap.

Benefits and work setup - Remote-first work in Germany with regular team sessions in Berlin and occasional sessions in Frankfurt and Munich, 30 days of vacation, equipment and learning support, and room to develop expertise across remote sensing and geospatial analysis.