Computer Vision Engineer, Senior

Responsibilities

• Architect and implement the entire embedded CV pipeline using high-performance Python and C++.
• Design and deploy robust multi-object tracking and sensor fusion algorithms for high-fidelity state estimation of fast-moving targets.
• Utilize real-time models (e.g., YOLO) and CNNs optimized for speed.
• Optimize code for low-latency performance on resource-constrained NVIDIA Jetson environments using TensorRT and TFLite.
• Apply principles of calibration, rectification, and 3D geometry to translate 2D footage into accurate 3D coordinates for fire control.
• Collaborate with robotics teams to ensure reliable perception data for autonomous decision-making.

Requirements

• Strong background in C++ (for performance) or Python.
• Deep expertise in VIO, VO, and SLAM frameworks for real-time state estimation.
• Proficiency in OpenCV and image-processing modules in PyTorch or TensorFlow.
• Experience with DNN inference libraries (e.g., TensorRT, TFLite) for edge deployment.
• Hands-on experience building embedded CV systems for real-time applications involving small, fast objects.
• Deep understanding of camera calibration and 3D geometry.

Preferred Qualifications

• Prior defense startup experience.
• Security clearance or ability to obtain one.