Machine Learning Wildfire Scientist

This is a fully remote position, open to applicants in United States.

📋 Description

• Lead EmberWorks' initiatives in fire behavior modeling, overseeing integration of new methodologies.

• Assess current fire behavior modeling strategies to determine the most suitable options for EmberWorks' data.

• Develop and execute models within an automated production framework utilizing airborne thermal observations alongside external fuel, weather, and terrain information.

• Investigate and innovate new modeling techniques through empirical, physical, and machine learning approaches as deemed appropriate.

• Take ownership of model validation and interpretation by establishing quality benchmarks, identifying erroneous outputs, and ascertaining whether problems arise from the model or upstream data.

• Specify computational needs for both training and inference processes.

• Collaborate with the platform engineer to set up GPU and distributed computing capabilities, ensuring an effective remote workflow.

• Foster relationships with the academic fire science community.

• Assist EmberWorks in engaging with research collaborators.

• Influence modeling strategies, validation techniques, and the methods by which outputs are delivered to users.

⛳️ Requirements

• Profound knowledge of fire behavior models including ELMFIRE, WRF-SITE, and FSIM, with a solid grasp of the underlying physics and mathematics.

• Practical experience in developing and training machine learning or data-driven models on spatiotemporal datasets.

• Strong background in remote sensing, particularly with raster imagery; thermal experience is a significant advantage.

• Expertise in scientific geospatial programming languages, including Python, xarray, GDAL, rasterio, PyTorch, or similar technologies.

• Familiarity with Git-based version control systems.

• Experience in conducting model training on cloud computing infrastructure, particularly utilizing GPU resources.

• Capability to work autonomously within a small R&D team while articulating modeling decisions to engineers and operations personnel.

• Bachelor's degree in fire science, atmospheric science, remote sensing, computer science, environmental science, or a closely related discipline.

• Proven track record of successfully delivering at least one wildfire project from inception to completion, or providing sustained, in-depth support for an established wildfire product suite.

• Active connections to academic or governmental fire research organizations.

• Experience in data assimilation or physics-informed machine learning techniques.

• Understanding of optics and radiometry, including thermal sensor calibration.

• Familiarity with wildland fire operations and the associated agencies and tools that facilitate these efforts.

• Advanced degree and research experience in a relevant field, or equivalent hands-on experience.

🏝️ Benefits

• Direct access to leadership.

• Opportunity to set technical direction.

• Work tested in real-world fire scenarios with immediate field feedback.

• Founding position on a small team engaged in pioneering efforts with unique airborne fire observation data.