Machine Learning Engineer (Semantic Scene Understanding)
Core
Design and deploy machine learning algorithms for real-time semantic scene understanding from UAV imagery to enhance operator situational awareness in defense operations.
Role type
Machine Learning Engineer (Semantic Scene Understanding)
Builds
Real-time semantic segmentation, object detection, and tactical feature extraction pipelines for UAVs
Domain
Defense / Autonomous Systems / Computer Vision
Deliverable
production ML models
Required skills
Python, PyTorch, C++, TensorRT, ONNX Runtime, CUDA, semantic segmentation, object detection, 3D-geometry algorithms, linear algebra, edge optimization
Preferred skills
PhD in CS/ML, experience with EO/IR imagery, experience shipping CV/ML to embedded systems
Technologies
PyTorch, TensorRT, ONNX Runtime, CUDA
Responsibilities
Develop state-of-the-art ML algorithms for semantic segmentation, object detection, and classification tailored to aerial imagery; Build high-level tactical features such as real-time road vectorization and dynamic obstacle mapping; Architect pipelines to align semantic data from multiple moving UAVs into a Common Operational Picture; Optimize and deploy algorithms into tactical C2 platforms using quantization and hardware acceleration
Seniority
Mid-Senior, hands-on IC
