SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Vibrant Planet is seeking an ML Engineer to build and operationalize foundation model-based deep learning systems that estimate forest structure metrics from remotely sensed data. The role sits at the intersection of machine learning, remote sensing, and forest ecology, supporting the company's Land Tender decision-support platform used by land managers, utilities, insurers, and community risk managers to lower wildfire risk.
Key responsibilities include adapting and fine-tuning geospatial foundation models with custom deep neural network heads to estimate forest metrics like canopy height, biomass, and basal area. You will prepare and curate training datasets from remote sensing sources (Sentinel-2, Sentinel-1, Landsat, lidar, NAIP) and field plot inventories, then evaluate model performance using standard remote sensing accuracy metrics and field-based validation.
On the infrastructure side, you will integrate trained ML models into Vibrant Planet's automated geospatial data pipeline as containerized, orchestrated inference services. This includes building and maintaining STAC (SpatioTemporal Asset Catalog) infrastructure, designing larger pipelines with Airflow DAGs, and ensuring idempotency, observability, and fault tolerance. You'll monitor pipeline health, implement model drift detection, and develop model cards documenting methods and performance.
The role emphasizes cross-team collaboration between SciDev, Data Engineering, and Product teams. You will contribute to scientific manuscripts, document pipelines and architectures, participate in code reviews, and translate requirements between scientific and engineering audiences. Strong organizational skills, self-motivation for remote work, and excellent adaptive communication are essential.
Required qualifications include an M.S. in Computer Science, Machine Learning, Remote Sensing, Data Science, Ecology, or related quantitative field (or equivalent work experience), plus 3+ years developing, training, and deploying deep learning models with PyTorch. You need strong Python proficiency (NumPy, pandas, xarray, scikit-learn), 3+ years with geospatial data processing tools (rasterio, GDAL, geopandas, shapely), and experience with workflow orchestration (Airflow, Prefect, Dagster). Git, GitHub, Docker, and CI/CD proficiency are required.
