Geospatial ML Engineer — Remote, Equity & Unlimited PTO

7Seventy; United States; Hybrid; USD 100,000 - 200,000; Full time

Benefits offered by this job

Health insurance

401(k) plan

Unlimited PTO

Company equity

Cell phone stipend

One-time home office setup allowance

Job summary

7Seventy is seeking an ML Engineer to build and operationalize foundation-model based deep learning systems that estimate forest structure metrics from remotely sensed data. You will fine-tune geospatial foundation models and develop custom neural network heads to process satellite and lidar data.

The role involves assembling training data from Sentinel-2, Sentinel-1, Landsat, lidar, NAIP, and field plots, evaluating with remote-sensing metrics, and integrating models into automated production

Qualifications

• 3+ years developing, training, and deploying deep learning models.
• 3+ years processing geospatial data.
• Experience with workflow tools (Airflow, Prefect, Dagster, or equivalent).
• Experience with containerization using Docker and cloud platforms (AWS preferred).

Responsibilities

• Adapt and fine-tune geospatial foundation models for domain-specific heads to estimate canopy height, biomass, and basal area.
• Prepare and manage training datasets from Sentinel-2, Sentinel-1, Landsat, lidar, NAIP, and field plots.
• Evaluate model performance using remote sensing metrics and field validation data.
• Integrate trained models into automated geospatial data pipelines as containerized inference services.
• Build and maintain STAC infrastructure and production Airflow DAGs.
• Maintain data ingestion, preprocessing, and quality-control workflows for satellite imagery.

Skills

Deep learning

Geospatial data

Python

PyTorch

Airflow

MLOps

Git

Education

Master's degree in Computer Science, Machine Learning, Remote Sensing, Data Science, Ecology, or related field

Tools

Docker

Kubernetes

PostgreSQL/PostGIS

STAC

Airflow

Job description

7Seventy is seeking an ML Engineer to build and operationalize foundation-model based deep learning systems that estimate forest structure metrics from remotely sensed data. You will fine-tune geospatial foundation models and develop custom neural network heads to process satellite and lidar data.

The role involves assembling training data from Sentinel-2, Sentinel-1, Landsat, lidar, NAIP, and field plots, evaluating with remote-sensing metrics, and integrating models into automated production