Data Scientist

Kuva Space is on a mission to solve the world's most pressing issues, such as climate change, food security, safety and security, by building the world's most extensive hyperspectral microsatellite constellation and developing AI-driven analytics services. We deliver reliable and timely global insights and foresight that transform rich spaceborne data into actionable insights customers can use for efficient resource management, optimizing operations, and improving profitability sustainably. Do you have strong Python engineering skills, hands-on machine learning experience, and fluency with geospatial data and tools? Would you like to work with technology that addresses environmental challenges and real-world problems? We are seeking a Data Scientist to design, build, and deploy machine learning solutions that extract actionable insights from hyperspectral satellite imagery. You will apply computer vision and remote sensing techniques to develop models that power core geospatial products and analytics. We offer you a challenging and meaningful role where you can make an impact and help organizations make better decisions. This is a full-time and permanent role. You can be based at our HQ in Espoo, Finland, or be fully remote. Fully remote employees should be within a maximum of 3 hours of the Eastern European Timezone.

Key Responsibilities

Research, prototype, and productionize ML/DL models for remote sensing use cases (e.g., classification, detection, segmentation, regression, anomaly detection) with hyperspectral satellite data Develop computer vision algorithms for imagery pre-processing, feature extraction, and quality control (radiometric, geometric, atmospheric considerations) Build robust Python codebases and data pipelines for training, evaluation, and inference; write clean, typed code with unit/integration tests and participate in code reviews Curate and manage datasets: ingestion, tiling, augmentation, labeling strategies, spectral feature engineering, and metadata management Analyze and process geospatial data using GDAL/Rasterio, GeoPandas, QGIS, and PostGIS; design efficient SQL queries for large geospatial datasets Optimize model performance and scalability on CPU/GPU, including batch inference, quantization/pruning when appropriate, and distributed training (e.g., Dask/Spark/PyTorch Distributed) Collaborate with product, engineering, and domain experts to translate requirements into well-scoped experiments and production-ready solutions Contribute to MLOps practices: experiment tracking, model/version management, CI/CD for ML, containerization, and cloud-based deployment Monitor model health and data drift; develop internal tooling for data validation, spectral calibration checks, and automated QA Document methods, datasets, and model behavior; stay current with advances in ML, computer vision, and remote sensing

Qualifications

Master’s degree in Computer Science, Data Science, Remote Sensing/Geoinformatics, Electrical/Computer Engineering, Physics, or related field; or equivalent experience Strong proficiency in Python and the scientific stack (NumPy, Pandas, SciPy, PyTorch), with experience building maintainable, tested codebases (type annotations, packaging, linting) Hands-on experience developing and deploying ML/DL models using scikit-learn and one or more deep learning frameworks (PyTorch or TensorFlow); experience with OpenCV Demonstrated experience with remote sensing imagery and geospatial tooling: GDAL/Rasterio, GeoPandas, QGIS; solid understanding of projections, georeferencing, and raster/vector data Proficiency in SQL and experience with PostgreSQL/PostGIS for spatial queries and performance tuning Familiarity with cloud or high-performance computing environments (e.g., AWS/GCP/Azure, Slurm/HPC), containers (Docker), and Linux/bash Strong problem-solving skills, ability to design rigorous experiments, and communicate results clearly to technical and non-technical stakeholders Experience working collaboratively in code (Git) and participating in peer reviews Excellent English language skills You share the company mission of improving life on Earth through daily, space-borne hyperspectral imaging and AI!🌎💚 Preferred qualifications Experience with hyperspectral imaging (sensor characteristics, spectral indices, unmixing, MNF/PCA, spectral libraries, anomaly detection) Experience with satellite or aerial data pipelines and photogrammetric/RS preprocessing (radiometric/atmospheric correction, orthorectification) Knowledge of geospatial data formats and standards: GeoTIFF/COG, STAC, Zarr, OGC APIs Experience with distributed data processing (Dask, Spark) and GPU acceleration; exposure to Kubernetes, serverless, or batch inference workflows MLOps tooling such as MLflow, Weights & Biases, or SageMaker; experience deploying models to APIs or streaming/tiling services Background in physics-based modeling, BRDF, or optical/radar/spectral remote sensing What do we offer? At Kuva Space, we offer a stimulating and safe work environment that encourages growth, collaboration, and excellence. Constantly learning new things is the norm here! Our focus on space technology means you'll have the opportunity to learn more about satellites, space, data, and the Earth. If you get excited about space-themed lunch table discussions, Kuva Space is the community for you 🪐✨ As an employer, we prioritize the well-being of our employees, both physically and mentally. Our health care benefits are comprehensive, including dental benefits and short-term psychotherapy. We offer an annual sports, culture and transport benefit. And our team members regularly meet up for after-work activities 🧘🏽 If you do not live in Finland but wish to, we can offer assistance in the migration process and support with learning the Finnish language and culture ⛄ In addition, you will be part of a dynamic, fun, and highly skilled team! Our diverse team of international colleagues, with varying cultures, shares a passion for deep tech and making our Earth more sustainable. And we believe Finnish 'sisu' is a must-have mindset to overcome any challenges that come our way ⛄ Are you excited to apply advanced machine learning and computer vision to geospatial challenges and build production-grade solutions? Then we’d love to hear from you. Please send us your application by the 14th of June. After that, we will begin reviewing applications and inform you of our decision as soon as possible.