Senior Data Quality Assurance Engineer, Data Annotation Services

Caterpillar Inc.•Irving, TX
•$112,710 - $183,140•Onsite

About The Position

Caterpillar is seeking a Senior Data QA Engineer to own the quality of training data used for robotics, autonomy, computer vision, and Vision-Language-Action (VLA) systems within a scrum team. The role combines data quality engineering, test automation, and machine learning to define, measure, and continuously improve the accuracy of annotated and captioned datasets from real-world heavy equipment and construction environments. The engineer will build automated validation frameworks and quality metrics, partner with annotation providers, ML engineers, and platform teams, and evaluate automated labeling and captioning technologies such as NVIDIA Cosmos Curator and Cosmos Reason. This hands-on technical position focuses on creating scalable, automated, production-grade quality assurance solutions. The role is full-time onsite at Caterpillar’s Irving, Texas (Dallas) office, five days a week. Domestic relocation assistance and visa sponsorship are available, subject to eligibility.

Requirements

  • Bachelor’s degree in Computer Science, Software Engineering, Data Science, Artificial Intelligence, Machine Learning, Robotics, or a related technical field.
  • Experience developing production-grade Python software for data validation, test automation, and data processing using modern software engineering practices, including version control, code review, CI/CD, Docker, and Kubernetes.
  • Experience designing and implementing automated data quality checks and testing frameworks for data pipelines, APIs, and datasets using tools such as Great Expectations, pytest, or similar technologies.
  • Working knowledge of machine learning evaluation methodologies and metrics, including precision, recall, F1 score, IoU, mAP, and error analysis.
  • Strong analytical and problem-solving abilities, exceptional attention to detail, and ability to communicate findings effectively across technical teams.
  • At least 2 years of experience working with camera and LiDAR sensor datasets for perception or machine learning applications, including data validation, ground truth generation, and data collection quality processes.
  • At least 3 years of experience in data annotation and quality assurance supporting autonomous systems or computer vision projects, including developing annotation guidelines and executing quality improvement initiatives.
  • Experience using annotation and dataset management platforms such as Labelbox, Supervisely, RoboFlow, or similar solutions, including QA and review workflows.
  • Experience with SQL and data analysis tools such as pandas and Jupyter notebooks to profile datasets, investigate defects, and create quality reporting.

Nice To Haves

  • Experience with computer vision, robotics, autonomy, Physical AI, or multimodal AI systems.
  • Experience validating AI-assisted annotation workflows, auto-labeling systems, and machine learning data pipelines.
  • Experience with machine learning frameworks and experiment tracking tools such as PyTorch, TensorFlow, ONNX, Weights & Biases, or MLflow.
  • Experience with NVIDIA Cosmos Curator, Cosmos Reason, or similar AI automation platforms.
  • Experience working with Vision-Language-Action (VLA) datasets and caption quality evaluation methodologies.
  • Experience with camera, LiDAR, radar, and autonomous machine sensor data.
  • Experience with cloud platforms, data observability tooling, MLOps, and data pipeline orchestration technologies.
  • Experience building human-in-the-loop quality workflows, inter-annotator agreement frameworks, and large-scale quality automation systems.

Responsibilities

  • Define, document, and maintain annotation guidelines, acceptance criteria, and quality standards for perception and VLA datasets.
  • Design and execute QA strategies for human, automated, and AI-assisted labeling and captioning, including sampling plans, audits, and golden-set validation.
  • Build automated data validation and test frameworks to verify labeling accuracy, consistency, completeness, schema conformance, and sensor synchronization across camera and LiDAR datasets.
  • Develop quality metrics, dashboards, and alerting capabilities, including precision/recall, IoU, inter-annotator agreement, and defect rates, to track dataset health and drive measurable quality improvements.
  • Identify, categorize, and root-cause labeling defects, working with annotation vendors, ML engineers, and domain experts to resolve issues and improve quality.
  • Build and validate quality gates within data pipelines for automated labeling, human review, quality validation, and dataset publication.
  • Evaluate and benchmark auto-labeling and auto-captioning platforms such as NVIDIA Cosmos Curator and Cosmos Reason, and help develop AI-driven QA capabilities that scale review processes.
  • Integrate QA workflows with Caterpillar data systems, external annotation providers, and human-in-the-loop processes while ensuring reliability, traceability, and observability.
  • Help establish and evolve quality standards for AI training data supporting computer vision, robotics, autonomy, and multimodal AI applications.

Benefits

  • Medical, dental, and vision benefits
  • Paid time off plan, including vacation, holidays, and volunteer time
  • 401(k) savings plans
  • Health Savings Account (HSA)
  • Flexible Spending Accounts (FSAs)
  • Health Lifestyle Programs
  • Employee Assistance Program
  • Voluntary Benefits and Employee Discounts
  • Career Development
  • Incentive bonus
  • Disability benefits
  • Life Insurance
  • Parental leave
  • Adoption benefits
  • Tuition Reimbursement
  • Domestic relocation assistance
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service