Egocentric Video Annotator

About the role - We are looking for detail-oriented Egocentric Video Annotators to support Physical AI data initiatives. You will work with first-person videos captured in industrial and commercial environments, converting real-world human activities into structured, high-quality datasets used to train and evaluate AI models.

Key responsibilities

• Perform temporal action segmentation by identifying activity start and end timestamps.
• Assign accurate action labels based on observed human activities.
• Write concise, descriptive captions for actions, objects, and hand movements.
• Identify left-hand and right-hand actions, object interactions, and task progression.
• Follow annotation guidelines and maintain consistency across video segments.
• Review annotations for timestamp accuracy, completeness, and label correctness.
• Meet daily productivity and quality targets while adhering to project SOPs.
• RequirementsExperience in video annotation, image labeling, or other data annotation projects is preferred.
• Strong attention to detail and the ability to observe activities frame by frame.
• Good written English and the ability to describe actions clearly and accurately.
• Basic understanding of timestamps, action sequences, and object interactions.
• Ability to follow detailed guidelines and work with structured annotation tools.
• Familiarity with industrial, manufacturing, automotive, or assembly workflows is an advantage.
• Immediate joiners preferred.