SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Overstory is building AI-powered solutions to help electrical utilities prevent catastrophic wildfires and power outages by analyzing vegetation risks using satellite imagery and advanced data science. The company uses cutting-edge technology to enable a resilient electrical grid in the face of climate change.
As Staff Data Scientist for Wildfire, you will lead the scientific foundation of Overstory's Fuel Detection Model—the core engine that translates satellite and environmental data into understanding of vegetation structure, fuel loads, and wildfire risk. This is a high-impact individual contributor role where you'll define model accuracy standards, design rigorous validation methodologies, and ensure modeling choices are grounded in fire science and remote sensing fundamentals.
Key responsibilities include: leading research into how vegetation structure and wildfire risk can be estimated from satellite, LiDAR, and environmental data across diverse geographies; designing validation and evaluation methodologies including ground-truth strategies and uncertainty quantification; prototyping and refining ML modeling approaches before handing off to ML engineers for production; integrating established fire science frameworks with data-driven methods; defining scientific standards for experimentation and reproducibility; communicating research findings to engineers, product teams, customers, and the wildfire science community; and mentoring ML engineers on scientific methodology.
You'll work alongside ML engineers and product teams in a remote-first, distributed environment spanning the Americas and Europe. The role requires deep expertise in remote sensing, geospatial analysis, statistical modeling, and machine learning, with 8+ years of applied research or data science experience in wildfire science, fire ecology, forestry, remote sensing, or related quantitative fields. Strong Python skills and familiarity with tools like GeoPandas, scikit-learn, PyTorch, and XGBoost are essential. Nice-to-haves include fire behavior modeling frameworks, physics-based ML integration, peer-reviewed publications, and cloud platform experience (GCP, Vertex AI).
