01Overview
Key Responsibilities
- Develop and deploy end-to-end Computer Vision and AI solutions for industrial inspection applications.
- Translate customer quality requirements into measurable AI performance targets.
- Select appropriate Computer Vision models, image-processing techniques, and inference strategies.
- Work with Automation Engineers to define cameras, lenses, lighting, triggering, and computing requirements.
- Manage image capture, dataset curation, annotation, training, validation, and model evaluation.
- Guide junior engineers on data collection, annotation, model training, testing, and performance evaluation.
- Integrate and troubleshoot Computer Vision models in Linux-based edge environments.
- Commission AI and Computer Vision systems at customer manufacturing sites.
- Troubleshoot image quality, hardware, model performance, inference, and production issues.
- Collaborate with software, electrical, mechanical, automation, and customer teams.
- Monitor deployed model performance and manage retraining, validation, releases, and rollbacks.
- Maintain documentation for datasets, experiments, model versions, evaluations, and deployments.
- Contribute to reusable Computer Vision architectures, standards, and best practices.
Required Qualifications
- Bachelor's or Master's degree in Computer Science, Software Engineering, or equivalent.
- 4 7 years of relevant experience in Computer Vision / Vision AI development and deployment.
- Robust knowledge of:
- Image Processing
- Image Classification
- Object Detection
- Image Segmentation
- Anomaly Detection
- Model Evaluation
- Strong proficiency in Linux and production edge environments.
- Working knowledge of Python for model development, data preparation, experimentation, and automation.
- Practical knowledge of industrial cameras, lenses, lighting, triggering, and image acquisition systems.
- Understanding of manufacturing processes, production lines, and automated quality inspection.
- Willingness to travel to customer sites in India and internationally. .
