Senior AI Trainer - Remote Video Annotation Expert

RemoteJobsOne; Hong Kong; Remote; HKD 454,000 - 757,000; Full time

Job summary

RemoteJobsOne is seeking a Senior AI Trainer for a fully remote contractor role. You will collaborate on a customer-driven project to enhance AI system accuracy through expert-level video annotation and data labeling.

Your work helps shape how models learn and perform using high-quality real-world input. Responsibilities include reviewing robotic-arm footage, tagging events with precise timestamps, and ensuring timestamp accuracy across datasets.

Qualifications

• Exceptional attention to detail and accuracy in reviewing and tagging visual data.
• Strong written and verbal communication skills for effective collaboration, reporting, and updates.
• Demonstrated time management and self-organization abilities for independent, remote project participation.
• Analytical and problem-solving mindset for annotation challenges.

Responsibilities

• Review video footage of robotic arms executing tasks, identifying key actions and outcomes with precision.
• Apply detailed grading guidelines to tag and annotate events, providing structured observations to improve model performance.
• Mark precise timestamps for start and end of key actions, ensuring frame-accurate segmentation of each video session.
• Verify timestamp accuracy and alignment across annotations, correcting drift or inconsistencies.
• Utilize video annotation tools to record findings and submit annotated datasets according to project standards.
• Collaborate with project trainers to resolve ambiguities, refine guidelines, and ensure annotation consistency.

Skills

Attention to detail

Communication skills

Time management

Analytical thinking

Job description

RemoteJobsOne is seeking a Senior AI Trainer for a fully remote contractor role. You will collaborate on a customer-driven project to enhance AI system accuracy through expert-level video annotation and data labeling.

Your work helps shape how models learn and perform using high-quality real-world input. Responsibilities include reviewing robotic-arm footage, tagging events with precise timestamps, and ensuring timestamp accuracy across datasets.