Senior Computer Vision Engineer

About the Role

As a Senior Computer Vision Engineer at Naveera, you'll develop the computer vision systems that power our AI dash cameras, fleet safety insights, and real-time video intelligence. From multi-camera calibration and object tracking to embedded AI deployment and GPU inference optimization, you'll turn high-resolution camera data into reliable perception systems that help NEMT and fleet operators run safer, smarter operations.

Responsibilities

• Design and develop computer vision systems for AI dash cameras and in-vehicle camera platforms
• Build real-time perception pipelines for road-facing and cabin-facing camera streams
• Develop models for object detection, multi-object tracking, driver monitoring, safety event detection, and road scene understanding
• Work on multi-camera calibration, including intrinsic and extrinsic calibration
• Apply stereo vision, multi-view geometry, 3D reconstruction, triangulation, and geometric projection models
• Build high-resolution video stream processing systems for real-time tracking at up to 60 FPS
• Optimize computer vision models for embedded AI chip platforms and edge deployment
• Improve GPU inference performance, latency, memory usage, and reliability
• Implement Kalman filtering, temporal fusion, multi-camera spatial fusion, and optimization methods for real-world camera systems
• Collaborate with embedded, firmware, backend, product, and operations teams to bring computer vision features into production

Required Qualifications

• Strong proficiency in Python and computer vision development
• Experience with PyTorch, OpenCV, TensorFlow, or similar ML and computer vision frameworks
• Strong understanding of multi-camera systems, intrinsic and extrinsic calibration, stereo vision, and multi-view geometry
• Experience with object detection, image segmentation, multi-object tracking, and real-time video analytics
• Experience building low-latency perception pipelines for production or real-world environments
• Understanding of GPU inference optimization, model deployment, and performance profiling
• Experience with geometric projection models, 3D reconstruction, triangulation, Kalman filtering, and temporal fusion
• Ability to evaluate computer vision models under real-world conditions such as motion blur, vibration, occlusion, lighting changes, and camera noise
• Strong engineering discipline with experience writing clean, testable, production-ready code

Preferred Qualifications

• Experience with embedded AI deployment and edge inference systems
• Familiarity with YOLOv8, YOLOv11, ByteTrack, DeepSORT, Segment Anything, or similar computer vision tools
• Experience with real-time camera pipelines, embedded Linux, GStreamer, CUDA, TensorRT, ONNX, or hardware-accelerated inference
• Familiarity with deploying models under compute, memory, thermal, and power constraints
• Experience using Git and working in collaborative engineering environments

Education / Experience Requirement

• PhD in Computer Vision, Robotics, Machine Learning, Electrical Engineering, Computer Science, or a related field, plus 1–2 years of relevant experience
• Master's degree in Computer Vision, Robotics, Machine Learning, Electrical Engineering, Computer Science, or a related field, plus 5+ years of relevant experience
• Bachelor's degree or equivalent practical experience, plus 7+ years of professional computer vision or perception engineering experience