Senior Architect, Computer Vision
Rockwell Automation14 days ago
Mayfield Heights, OH, USA or Milwaukee, WI, USASenior
Responsibilities
- Own model architecture, training, evaluation, and deployment approaches for FactoryTalk Analytics VisionAI and vision capabilities across the FactoryTalk portfolio.
- Develop dataset strategy, annotation pipelines, model selection, fine-tuning, quantization, and inference deployment for edge hardware.
- Extend evaluation suites and release gates while measuring and monitoring accuracy, drift, and failure modes in customer deployments.
- Review designs, coach engineers, and serve as a recognized computer vision and vision-language models resource across engineering teams.
- Partner with Product Management and fellow architects on portfolio roadmaps and cost, latency, and accuracy tradeoffs.
- Design for on-premises and disconnected inference, model provenance and signing, data residency, and protection of customer and Rockwell intellectual property.
- Represent Rockwell’s industrial vision perspective in customer engagements, partner technical reviews, and external technical forums.
Requirements
- Bachelor's Degree or equivalent years of relevant work experience.
- Legal authorization to work in the United States is required.
- Typically requires a minimum of 8 years of related experience.
- Production experience training, evaluating, and deploying computer vision models beyond research or prototype settings.
- Hands-on experience with modern vision or vision-language model architectures and PyTorch or TensorFlow.
- Familiarity with Industrial Control Systems.
- Experience applying vision-language models to visual question answering, scene or document understanding, grounded instruction following, or multimodal retrieval.
- Experience with edge AI platforms and inference optimization, including NVIDIA Jetson, TensorRT, quantization, pruning, or distillation.
- Experience with industrial or safety-relevant vision applications such as quality inspection, defect classification, robotic guidance, OCR on production surfaces, or safety monitoring.
- Experience with synthetic data generation or simulation-based training, including digital twin environments and domain randomization.
- Experience establishing MLOps practices for vision work, including dataset versioning, reproducible training, model registries, and automated evaluation gates.
- A Master's or PhD in Computer Vision, Machine Learning, Robotics, or a related discipline is preferred.
- Published research, open-source contributions, or conference presentations in computer vision or multimodal machine learning are preferred.
Benefits
- Hybrid work in Mayfield Heights, Ohio, or Milwaukee, Wisconsin.
- Employees are expected to work at a Rockwell location at least Mondays, Tuesdays, and Thursdays unless they have a business obligation outside the office.
- Health insurance including medical, dental, and vision coverage.
- 401(k).
- Paid time off.
- Parental and caregiver leave.
- Flexible work schedule coordinated with the manager.
Tech Stack
Categories