株式会社SolafuneのFull Stack Engineer | Satellite Imagery & Geospatial AIの求人

▼About Us
We're a fast-paced startup harnessing AI to unlock insights from satellite and earth observation data. As an early team member, you'll have a significant impact on shaping our products and driving innovation.

▼About the Role
We're looking for a mid-level engineer who blends machine learning expertise on satellite imagery with the ability to ship full-stack applications. You'll wear multiple hats: training models, building the pipeline, and deploying them next. This is a hands-on role for someone who thrives in ambiguity and wants to build things end-to-end.

▼What You'll Do
・Design, train, and deploy ML/deep learning models (CNNs, vision transformers, segmentation, object detection) on satellite and aerial imagery.

・Work with multispectral, Earth Observation, and high-resolution optical data — including preprocessing, georeferencing, and tiling pipelines.

・Build back-end services (APIs, data pipelines, model serving) that power our geospatial AI features.

・Develop front-end interfaces to visualize imagery and model outputs (interactive maps, dashboards).

・Help shape the technical direction and collaborate with the research team to identify areas for improvement and inform product development.

・A startup mindset: self-directed, adaptable, and comfortable owning projects end to end

・Familiarity with cloud platforms (AWS, GCP, or Azure) and Docker

・3–6 years of experience, with demonstrated work applying ML/DL to satellite or remote sensing imagery

・Strong Python skills and hands-on experience with PyTorch or TensorFlow

・Experience with geospatial tooling (GDAL, rasterio, GeoPandas, or similar)

・Full-stack ability: back end (Python/FastAPI, Flask, or Node.js) and front end (React/JavaScript with mapping libraries like Leaflet, Mapbox, or OpenLayers)

・Experience with earth observation sources (Sentinel, Landsat, Planet, Maxar)

・MLOps and model deployment at scale

・Familiarity with geospatial data standards and coordinate systems