Wildfire Perception Engineer - Real-Time ML/CV

Rain Industries Inc; Alameda (CA); On-site; USD 180,000 - 230,000; Full time

Job summary

Rain Industries Inc. seeks an experienced engineer to give autonomous aircraft the perception needed for wildfire suppression. You will fuse real-time visual, thermal, and geospatial data to create a dynamic understanding of fire fronts and environments.

Responsibilities include developing perception software, collaborating with field ops, and building tools for sensor data ingestion. Travel up to 15% for validation testing is expected.

Qualifications

• 6+ years of experience in computer vision and machine learning, including segmentation, localization, and world modeling.
• Strong proficiency in C++14 (or newer) development for resource limited environments.
• Experience designing algorithms for sensors (visual, thermal, RADAR, LiDAR).
• Strong mathematical skills and probabilistic modeling for uncertainty.

Responsibilities

• Apply ML and classical techniques to real-time mapping of wildfire environments onboard aircraft.
• Drive design, development, testing, and deployment of the perception system with engineering and field ops.
• Produce performant software for ingesting sensor data and building wildfire environment models.
• Design experiments and curate training/evaluation sets.
• Architect and implement core product apps, tools, and infrastructure.

Skills

Computer Vision

Machine Learning

C++

Sensor Fusion

Probabilistic Methods

Team Collaboration

Education

MS or PhD in Robotics/CS/EE or related field

Tools

CUDA

Job description

Rain Industries Inc. seeks an experienced engineer to give autonomous aircraft the perception needed for wildfire suppression. You will fuse real-time visual, thermal, and geospatial data to create a dynamic understanding of fire fronts and environments.

Responsibilities include developing perception software, collaborating with field ops, and building tools for sensor data ingestion. Travel up to 15% for validation testing is expected.