Middle Computer Vision Engineer (Motion-Aware Detection and Classification)

Key Responsibilities

• Develop, train, and optimize deep learning models for:
  • Motion detection under ego-motion conditions;
  • Object detection and classification;
• Design motion-aware preprocessing and compensation techniques (stabilization, optical flow, background subtraction)
• Perform model evaluation using quantitative metrics (mAP, precision/recall, latency, robustness under motion)
• Optimize models for deployment (TensorRT/ONNX/edge devices where applicable)

Required Qualifications

• 2–4 years of hands-on experience in Computer Vision;
• Practical experience training object detection models (e.g., YOLO, Faster R-CNN, RetinaNet);
• Understanding of motion-related CV concepts:
  • Optical flow;
  • Video tracking;
  • Ego-motion compensation;
  • Kalman filtering (basic level);
• Experience working with video streams (not only static images);
• Solid Python skills;
• Familiarity with OpenCV;

Nice to Have

• Experience with multi-sensor fusion (automotive/smart home);
• Experience optimizing models for edge deployment;
• Understanding of real-time systems constraints (latency, throughput, memory);
• Experience working with noisy, real-world datasets;