Remote Robot Video QC Reviewer — Annotation Quality

Welocalize, Inc.; Malaysia; Remote; MYR 28,000 - 55,000; Full time

Job summary

Welocalize, Inc. seeks a detail-oriented QC Reviewer to ensure the quality and consistency of robot manipulation video annotations.

You will review labeled segments against a thorough style guide, catch errors, and provide constructive feedback to annotators to maintain training data accuracy for robotics AI. Work is project-based and fully remote, with opportunities to participate in calibration sessions and expand expertise in robotics terminology and data labeling practices.

Qualifications

• Excellent written English and strong reading comprehension.
• Exceptional attention to detail to catch subtle mismatches between video content and written description.
• Strong grasp of spatial/mechanical description to verify annotator accuracy.
• Ability to apply a detailed, structured style guide consistently.

Responsibilities

• Review robot manipulation video annotations at three labeling levels (atomic motion, skill/subtask, task/goal).
• Cross-check annotator descriptions against all three camera views to verify accuracy of actions, objects, and grips.
• Verify full video coverage with no gaps in labeling, including idle moments.
• Flag discrepancies where descriptions misstate outcomes or are smoothed over.
• Ensure vocabulary and terminology align with the style guide and vary language to avoid repetition.
• Identify recurring errors and report patterns to the team lead for calibration.
• Provide actionable feedback referencing the style guide for rejected or revised annotations.
• Participate in calibration sessions with the team and client to align QC standards.
• Maintain a QC log documenting review outcomes and annotator trends.

Skills

Attention to detail

Written English

Spatial reasoning

Style-guide application

Constructive feedback

Independent work

Video annotation familiarity

Label Studio experience

Tools

Label Studio

Job description

Welocalize, Inc. seeks a detail-oriented QC Reviewer to ensure the quality and consistency of robot manipulation video annotations.

You will review labeled segments against a thorough style guide, catch errors, and provide constructive feedback to annotators to maintain training data accuracy for robotics AI. Work is project-based and fully remote, with opportunities to participate in calibration sessions and expand expertise in robotics terminology and data labeling practices.