Toloka Annotators - First Person AI Video Trainer

Data Collector is needed in Austin, United States.

Client: Toloka Annotators

Location: Austin, TX

Contract: Part-time

Job Description

This is a project-based opportunity on an AI training platform — not a job. No fixed hours, no commitment beyond what fits your schedule. You record, you get paid. We’re looking for people to record point-of-view videos of everyday household activities. You mount your smartphone on your head, go about your routine, and the camera captures exactly what you see. These videos help train AI systems and robots to understand how people interact with objects in real-world environments.

Do you wash dishes every day? Vacuum the rug? Walk the dog? Fold laundry? What if you could get paid just for recording it?

Responsibilities

• Record everyday household activities from a first-person perspective — cooking, cleaning, tidying up, folding laundry, gardening, walking the dog, and more
• Record in your own home, yard, garage, or workspace
• Follow short recording guidelines per session
• Upload videos through the project app

Requirements

• Equipment: Own or use a smartphone or alternative camera device capable of 30 FPS or greater and 1080p resolution or greater. iPhone 11 or newer, Google Pixel 6+, or Samsung Galaxy S21+ (must have a wide-angle / 0.5x lens)
• Capture method: Mount the device at forehead or eye-level for an egocentric perspective, maintaining a downward angle to maximize hand visibility.
• Software: Utilize the designated “D&A” application for all data capture and upload.
• Quality control: Ensure all video footage is clear, stable, well-lit, at least 15 seconds long, and accurately depicts useful, intentional tasks without unnecessary pauses or contrived movements.

Additional Information

This freelance opportunity allows you to earn while you do your household chores: up to $10 per hour equivalent depending on the project. This is project-based freelance work — pick up recording sessions when they fit your schedule. Influence the future of how AI and robots understand and interact with the real world.