Vision-Language-Action (VLA) Annotator

About the role

We are seeking detail‑oriented annotators to create high‑quality training data for Vision‑Language‑Action and robotics models. You will work remotely, reviewing videos of real‑world activities and producing precise action segmentations and descriptions.

Key responsibilities

• Watch videos, segment them into logical action‑based intervals with accurate start and end times, ensuring no gaps or overlaps.
• Review and edit AI‑generated captions, writing clear, concise descriptions based solely on visible actions, objects, tools, and locations.
• Maintain consistent terminology, flag unusual video portions, and ensure annotations meet project‑specific quality standards.
• Perform quality checks for segmentation, caption accuracy, grammar, and completeness before submission.
• Use annotation shortcuts and AI‑assisted tools efficiently while meeting productivity targets.

Required profile

• Strong attention to detail and ability to follow detailed written guidelines.
• Excellent written English with the ability to describe actions clearly and precisely.
• Good observational skills to distinguish subtle differences in actions, objects, and timing.
• Comfortable reviewing video content for extended periods and performing repetitive annotation work accurately.
• Basic computer proficiency, reliable internet connection, and ability to learn new annotation platforms quickly.
• Openness to feedback and ability to adapt to guideline changes.