Remote Geospatial ML Scientist - Satellite Change Detection

Turtle Trax S.A.; United States; On-site; USD 140,000 - 210,000; Full time

Job summary

Turtle Trax S.A. is seeking an experienced Data Scientist to develop geospatial ML solutions for satellite-based early-warning of wind-driven sand encroachment and pipeline displacement. You will own detection/calibration of alerts from raw satellite passes and validate against field-logged events.

The role emphasizes Python-based analysis, SAR/optical data fusion, and model validation using ROC/AUC metrics, with remote work options for a world-scale geospatial project.

Qualifications

• 7+ years applied data science / ML with geospatial remote sensing.
• Strong Python proficiency including numpy and common geospatial libs.
• Experience with deep learning and model validation metrics.

Responsibilities

• Build and calibrate change-detection and anomaly models on Sentinel-1/2 imagery.
• Learn per-site normal terrain baselines and validate detections with ground truth.
• Fine-tune geospatial foundation models with LoRA/PEFT on limited labeled data.
• Implement SAR techniques for displacement and dune-migration tracking.
• Ingest data from Copernicus and fuse optical, SAR, DEM, and wind data.
• Design labeling strategy and train/validation splits to avoid leakage.
• Communicate results and limitations to technical and business stakeholders.

Skills

Geospatial ML

Python

PyTorch

Education

MSc/PhD in Remote Sensing

Tools

PyTorch

Rasterio/GDAL

xarray

geopandas/shapely

Job description

Turtle Trax S.A. is seeking an experienced Data Scientist to develop geospatial ML solutions for satellite-based early-warning of wind-driven sand encroachment and pipeline displacement. You will own detection/calibration of alerts from raw satellite passes and validate against field-logged events.

The role emphasizes Python-based analysis, SAR/optical data fusion, and model validation using ROC/AUC metrics, with remote work options for a world-scale geospatial project.