Smart Ship Hub Digital
Singapore | Japan | India
We are looking for a skilled
Camera Vision Engineer (Video Processing Specialist) with hands-on experience in Deep Learning, video analytics, and edge device deployment. The ideal candidate will work on real-time camera vision pipelines, sensor fusion systems, and video analytics applications across industrial and maritime domains.
Key Responsibilities
- Develop and optimize real-time video analytics solutions using DeepStream, YOLO, PyTorch, and OpenCV.
- Build and deploy computer vision models for object detection, segmentation, tracking, OCR, pose estimation, and classification.
- Design and integrate sensor fusion pipelines for enhanced situational awareness and decision-making.
- Work with RTSP/industrial cameras, satellite imagery, and multi-camera setups for real-time processing.
- Deploy solutions on edge devices (Jetson series) and integrate via Docker, Linux, and GStreamer.
- Build user-friendly dashboards and interfaces using PyQt and integrate with backend platforms.
- Handle WebRTC, Kafka, Zookeeper, PostgreSQL/MySQL for scalable system architecture.
- Collaborate with cross-functional teams to deliver high-quality CV solutions and ensure system reliability.
- Participate in requirement gathering, project management, testing, optimization, and deployment.
Required Skils
- Deep Learning (CNNs, YOLO, DETR v2, SAM2, OCR)
- PyTorch, TensorRT, ONNX Runtime
- OpenCV, DeepStream, GStreamer
- Video Processing & Action Recognition
- Jetson/Edge Devices
- Docker, Linux, Git
- Kafka, Zookeeper
- PostgreSQL, MySQL
- Python, C++, Bash Scripting
Experience: 2-4 Years
Additional Skills
- Experience with ROS, PLC integration
- Satellite image analysis (preferred)
- GUI development using PyQt
- Familiarity with cloud platforms (Azure/EC2
Qualifications- B.Tech/B.E in Computer Science, Electronics, AI/ML, or related field.
- Certifications preferred:
- NVIDIA Jetson AI Specialist
- Microsoft Azure (AZ-900, AZ-104, AI-900)
- Published research or hands-on project portfolio is a plus.
