Machine Learning (ML) Vision Engineer Location: Remote (occasional onsite in Middletown, PA) Client: Pennsylvania Turnpike Commission (PTC) State of PA Engagement Type: Contract Project: Unified Back Office System (UBOS) Initiative C2C Visa: USC, H1B Role Overview The Pennsylvania Turnpike Commission (PTC) is seeking a Machine Learning (ML) Vision Engineer with strong software engineering expertise and proven experience in computer vision, image processing, and machine learning. This position is central to the Optical Character Recognition (OCR) initiative that powers real-time, large-scale image-based transaction processing within the UBOS platform. The ideal candidate combines deep technical knowledge of ML Vision systems with the ability to deliver production-grade, scalable, and reliable code in a distributed environment.

Key Responsibilities

• Design, implement, and optimize ML-based computer vision solutions, including: Object detection and image classification models. OCR pipelines integrating third-party and custom-built OCR engines. Image quality assessment classifiers (e.g., overexposed, underexposed, obstructed).
• Develop high-quality, production-ready Python code using object-oriented design (OOD) and clean architecture principles.
• Collaborate with system architects to integrate ML and vision components into a distributed, message-driven system architecture.
• Build data-driven training, testing, and feedback loops for OCR and vision models.
• Optimize ML models for low-latency and high-throughput performance in real-time transaction workflows.
• Partner with database, API, and DevOps teams to ensure seamless deployment, testing, and model lifecycle management.
• Participate in system-level design reviews and architecture discussions to ensure scalability and maintainability.
• Develop frameworks and pipelines suitable for production use—beyond proof-of-concept prototypes.
• Produce comprehensive technical documentation, including design specifications, test plans, system diagrams, and lessons learned.
• Collaborate with cross-functional teams, stakeholders, and project sponsors throughout the software development lifecycle.
• Provide onsite support during post go-live phases when required.

Minimum Qualifications

• Strong professional experience with Python (mandatory).
• Deep understanding of software engineering principles: OOP, design patterns, and clean code architecture.
• Hands-on experience with image processing and computer vision frameworks (e.g., OpenCV, Pillow).
• Experience with machine learning frameworks such as PyTorch or TensorFlow.
• OCR-related expertise using tools such as Tesseract, PaddleOCR, or EasyOCR.
• Familiarity with object detection architectures (YOLO, Faster R-CNN, SSD, etc.).
• Knowledge of classification models, feature extraction, and performance evaluation metrics for vision tasks.
• Proficiency with Microsoft Office 365 (Word, Excel, PowerPoint, Teams) and Microsoft ADO Testing Module.

Desired Skills & Attributes

• Proven track record in building ML Vision systems for production environments (not just research or proof-of-concept).
• Strong focus on code quality, maintainability, and scalability.
• Experience working within distributed or microservices-based architectures.
• Excellent analytical and problem-solving skills with an ability to think innovatively.
• Effective communication skills with the ability to explain technical concepts to non-technical stakeholders.
• Additional preferred experience: Exposure to Enterprise Business Solutions (e.g., SAP). Familiarity with ITIL / ITSM methodologies.

Education & Certifications

• Bachelor’s degree in Computer Science, Information Systems, or a related field (required). Equivalent combination of education and relevant experience may be considered.
• Certifications in Machine Learning, AI, or Computer Vision (preferred).