Sr. Software Engineer, Computer Vision

RESPONSIBILITIES:

• Lead computer vision model development and deployment for real-time inspection and automated defect recognition in production-scale manufacturing
• Architect and implement pipelines for scalable model training, evaluation, deployment, monitoring, and retraining using industry-standard tools like Ray Train/Serve, Kubeflow, Airflow, or equivalent
• Develop software to integrate hardware, sensors, and tooling (including sensor fusion) with AI-driven process monitoring systems
• Build and optimize data pipelines from production lines, incorporating real-time stream processing and inference
• Collaborate with part/process engineers, NDE engineers, and materials scientists to link advanced models and outputs to manufacturing
• Lead design and code reviews, technology evaluations, and enforce best practices (e.g., style, CI/CD, accuracy, testability, efficiency, and standards)
• BASIC QUALIFICATIONS:
• Bachelor's degree in computer science, engineering, math, physics, or related STEM discipline; OR 8+ years of professional experience building and deploying AI software/Machine Learning in lieu of a degree
• 5+ years of software development experience
• 3+ years deploying AI models to production environments
• 3+ years of software engineering experience
• 3+ years of experience leveraging Python for data analysis
• PREFERRED SKILLS AND EXPERIENCE:
• Proven track record as a technical leader in software projects involving AI/ML
• Hands-on experience deploying computer vision models for real-time manufacturing inspection and defect detection
• Strong development experience in Python and C++ (or similar), with expertise in ML frameworks like PyTorch or JAX
• Hands-on experience with computer vision libraries (e.g., OpenCV)Experience fine tuning and adapting state-of-the-art models, such as transformers and vision transformers, to production environments
• Experience fine-tuning LLMs or vision-language models and building agentic tools
• Experience deploying applications at scale with Docker, Kubernetes, and cloud/edge inference for factory automation