Data Scientist — Imagery Ingest & Data Platform

2) Data Scientist — Imagery Ingest & Data Platform

Location: Stockholm (hybrid) or remote in Sweden

Reports to: Geospatial Lead / Head of Data

Type: Full-time

Find someone with PHD in computer vision

Role summary

You’ll build and operate our imagery ETL: automated ingest from open/commercial sources, a searchable metadata catalog, cloud-optimized/tiled storage, and model-ready datasets. You own orchestration, reliability, data quality; handoff to MLEs for cost monitoring and production model serving.

What you’ll do

• Automate ingestion (e.g., Sentinel/Landsat and commercial feeds), harmonize provider metadata to STAC, and build a searchable STAC catalog.
• Implement preprocessing: cloud masking, mosaicking, pyramids/tiling, geospatial indexing (GeoParquet/PostGIS).
• Orchestrate pipelines with Airflow/Prefect; optimize compute/storage (S3/GCS, COGs, partitioning).
• Stand up reliable batch/online model serving with containers/Kubernetes and proper queuing/scaling.
• Monitor timeliness, availability, and costs; trigger alerts for anomalies/spikes.
• Ensure reproducibility (MLflow/Kubeflow/CI/CD), strong logging/observability, and clear data schemas/APIs to the platform.

Must-have

• 3+ years as a data engineer; hands-on EO preprocessing with rasterio/xarray preferred.
• Strong Python; experience with GCS, containers, and Kubernetes.
• Proven orchestration with Airflow/Prefect and production observability.
• Geodata modeling (PostGIS/GeoParquet) and tile-serving strategies.
• Practical MLOps exposure—delivered batch/online inference to real users.
• Manage a parallel compute framework (Spark or Ray) for large-scale imagery processing.

Nice-to-have

• Cost-optimized large-scale EO processing and commercial feed integrations.
• COG pipelines, STAC catalogs, multi-AOI/time-range caching strategies.
• Data security and access control for geospatial services.

Why Centersource

• You own the backbone that makes our models usable, scalable, and affordable.
• Direct business impact: your pipelines drive decisions, forecasts, and revenue.
• Product-centric environment focused on reliability, performance, and cost.

How to apply

Email your CV/LinkedIn and a concise description of a production EO/imagery pipeline you built to mai@centersource.io

Subject: Satellite Data Engineer.

Working language: English; Swedish is a plus. Applicants must have the right to work in Sweden.