Satellite and drone imagery access is on the rise, and traditional image processing methods are struggling to keep up. We’ve never had more data, and yet it’s never harder than ever to gain meaningful insights.
Our scalable AI platform enables custom model training on global features, providing real-time, on-demand geospatial insights with impressive speed and accuracy. The application turns months of manual work into mere minutes, and with much better results. We work with customers from various domains, from intelligence and defence, local and federal governments, to small and large enterprise enterprises, which requires us to have a lot of flexibility on how we deploy and maintain our services.
We kicked off in 2020 and have secured $35 million in series A funding from a lineup of top US and European investors, among which Microsoft M12, Point72 Ventures, Maxar, In-Q-Tel, SAFRAN, and ISAI/Capgemini.
We're searching for a Software Engineer to join our Data Plane team, where you'll build the preprocessing and postprocessing services that transform raw imagery into AI-ready inputs and convert model outputs into actionable geospatial products. You'll work on high-throughput data pipelines that handle terabytes of satellite, aerial, and drone imagery across diverse formats and coordinate systems.
What you'll do
- Build and optimize preprocessing pipelines that ingest, tile, and transform geospatial imagery (GeoTIFF, multispectral, SAR, COG) for downstream ML inference
- Develop postprocessing services that convert model outputs into production-ready deliverables: segmentation masks, probability maps, and vector detections in GeoPackage format
- Design resilient, memory-efficient services for processing large-scale imagery through distributed worker pools
- Work on our unified Huntr-to-Replika pipeline, automating the flow from detection outputs to 3D-ready terrain tiles with auto-generated configuration
- Tackle challenges around coordinate reference systems, spatial indexing, and data format interoperability
- Optimize for the unique characteristics of geospatial workloads: memory-bound processing, batch-heavy operations, and streaming large raster datasets
YOUR PROFILE
- Strong practical knowledge of Python with experience building data-intensive applications
- Experience with geospatial data formats, GDAL, or raster/vector processing
- Understanding of coordinate reference systems and spatial data transformations
- Experience with data pipelines, ETL processes, or batch processing systems
- Familiarity with async processing patterns, task queues (Redis), and worker architectures