We are seeking a Computer Vision Engineer with hands-on experience in PyTorch and TensorFlow to design, develop, and deploy deep learning–based vision systems for manufacturing automation. The ideal candidate will apply AI and image processing to improve product quality, automate inspections, and enhance process efficiency across the production floor.
Key Responsibilities
- Design and train deep learning models using PyTorch and/or TensorFlow for tasks such as defect detection, surface inspection, object localization, and classification.
- Build and optimize computer vision pipelines, including image preprocessing, data augmentation, and real-time inference.
- Integrate AI models with industrial systems—robotic cells, PLCs, and camera networks—for automated inspection and guidance.
- Deploy trained models to edge computing platforms (e.g., NVIDIA Jetson, Intel OpenVINO, or ONNX Runtime) for real-time inference.
- Collect, label, and manage datasets from production lines to improve model robustness and accuracy.
- Collaborate with automation, quality, and process engineering teams to define system requirements and deploy vision solutions in manufacturing environments.
- Monitor and maintain deployed models, implementing retraining and continuous improvement cycles.
- Research emerging techniques in computer vision and deep learning, staying current with innovations in PyTorch, TensorFlow, and model optimization.
Required Qualifications
- Bachelor’s or Master’s degree in Computer Science, Electrical/Mechanical Engineering, Robotics, or related field.
- 2+ years of hands-on experience with PyTorch and/or TensorFlow for computer vision applications.
- Strong understanding of CNNs, Vision Transformers, object detection models (e.g., YOLO, Faster R-CNN, SSD), and segmentation networks (e.g., U-Net, Mask R-CNN).
- Proficiency in Python and familiarity with supporting libraries: OpenCV, NumPy, scikit-learn, Pandas, Matplotlib.
- Experience deploying and optimizing models for real-time inference in manufacturing or industrial environments.
- Understanding of camera calibration, lighting, optics, and image acquisition systems.
- Familiarity with version control (Git), Docker, and Linux environments.
Preferred Qualifications
- Experience with 3D vision, depth cameras, or LiDAR for dimensional inspection.
- Experience integrating models with C++ or industrial vision platforms (Cognex, Keyence, Basler).
- Familiarity with ONNX, TensorRT, or OpenVINO for model conversion and acceleration.
- Knowledge of MLOps workflows for model deployment and monitoring.
- Experience in Industry 4.0, predictive maintenance, or smart manufacturing applications.