QA Reviewer, Data Annotation

Mind Robotics • Palo Alto, CA

3d

About The Position

Mind Robotics is building Physical AI for real-world industrial deployment, starting on the factory floor. The company is seeking a QA Reviewer & Calibration Lead to ensure its data annotation pipeline maintains high throughput without sacrificing quality. This role supports quality control for a continuous data flywheel by leading alignment sessions, identifying systematic labeling inconsistencies, and working with internal reviewers and partner teams to resolve ambiguous edge cases.

Requirements

• 2–4 years of experience in quality assurance, data annotation QA, or data review in high-velocity technology or AI environments.
• Meticulous attention to detail, including the ability to identify subtle edge cases and resolve ambiguous data scenarios logically and practically.
• Strong written and verbal communication skills, including the ability to explain complex labeling rules and facilitate alignment discussions.
• Comfort navigating dashboards, spreadsheets, and basic reports to track accuracy metrics and identify trends.
• A proactive, team-oriented, collaborative mindset focused on coaching reviewers and continually improving quality.

Nice To Haves

• Experience with multi-camera or ego-exo video annotation, point cloud data, or robotics training data pipelines.
• Experience working with third-party vendor teams (BPO or managed services) in a fast-paced hardware or AI deployment environment.

Responsibilities

• Lead weekly alignment meetings with internal reviewers and vendor partners to promote consistent grading and labeling.
• Identify common reviewer errors, conduct root-cause analysis on labeling discrepancies, and build feedback loops to eliminate systemic mistakes.
• Maintain clear and up-to-date labeling instructions, edge-case taxonomies, and standard operating procedures as project requirements evolve.
• Track quality scores, accuracy rates, and inter-annotator agreement metrics across dashboards to help ensure production targets are met.
• Provide direct feedback and targeted training to reviewers handling complex guidelines or multi-sensor data edge cases.
• Collaborate with technical leads, operations, and annotation teams to resolve taxonomy questions and support the data flywheel.

What This Job Offers

• Job Type: Full-time
• Career Level: Mid Level
• Education Level: No Education Listed