
Computer Vision Engineer
- India
- Permanent
- Full-time
- Design and implement computer vision models and systems tailored to business requirements using deep learning techniques.
- Develop object detection, image classification, segmentation, and tracking solutions using YOLO and other relevant architectures.
- Use and customize OpenCV and PyTorch for image preprocessing, augmentation, and model training.
- Optimize model performance for both accuracy and inference speed on edge devices and cloud environments.
- Work with large-scale datasets: data collection, cleaning, annotation, and augmentation.
- Perform rigorous validation and testing of CV models to ensure robustness and generalizability.
- Collaborate with cross-functional teams for integration of computer vision modules into products and platforms.
- Keep up-to-date with the latest research trends, tools, and libraries in computer vision and deep learning.
- 4-8 years of industry experience in developing and deploying computer vision applications.
- Proficiency in Python with hands-on experience in PyTorch for deep learning model development.
- Strong command of OpenCV for image and video processing tasks.
- Expertise in YOLO (You Only Look Once) for real-time object detection.
- Solid understanding of computer vision fundamentals including feature extraction, object detection, classification, and tracking.
- Experience in training, fine-tuning, and evaluating deep learning models on real-world datasets.
- Ability to write clean, efficient, and modular code with proper documentation.
- Familiarity with GPU acceleration and performance optimization for inference.
- Experience working with version control systems like Git.
- Master's or Bachelor's degree in Computer Science, Electrical Engineering, Artificial Intelligence, or a related field.
- Experience deploying CV models in production environments (e.g., on cloud or edge devices).
- Familiarity with ONNX, TensorRT, or similar model optimization frameworks.
- Knowledge of CI/CD pipelines for ML deployment.