The work
You will collect real-world video used to help train AI systems. Record yourself doing ordinary activities indoors, outdoors, while walking, and during vehicle navigation or driving-related tasks.
Use a third-person camera setup mounted above and behind you, angled slightly downward. Your full body, movement, and surrounding context must stay visible throughout each activity.
- Record continuous clips that are at least 60 seconds long.
- Keep the camera stable and capture the relevant activity and environment clearly.
- Securely mount the camera to a vehicle for navigation tasks.
- Follow recording, positioning, quality, and safety instructions.
- Upload footage for review and submit video that passes quality review and is accepted for delivery.
What it pays and takes
This is an entry-level, part-time contractor role. You need access to suitable recording equipment and must be comfortable handling video files and uploading footage for review.
- Pay: $15 to $30 per hour.
- Camera: Insta360, DJI Osmo 360, GoPro MAX, or an equivalent camera.
- Camera specifications: At least a 170-degree field of view and minimum 720p resolution.
- Recording setup: A stable third-person position that keeps your full body in frame.
- Activities: Indoor, outdoor, walking, and vehicle-based recording.
- Language: Fluent in English.
- Location: Open to candidates in AE, AR, AT, BD, BE, BH, BR, CA, CH, CL, CO, DE, DK, DZ, EG, ES, FR, GB, GH, GR, ID, IN, IT, JO, JP, KE, KG, KW, KZ, LB, LK, MA, MX, MY, NG, NL, OM, PE, PH, PK, PS, PT, QA, SA, SG, TH, TN, TW, US, UZ, VN, and ZA.
- Helpful experience: 360-degree video recording, action cameras, backpack or harness mounts, extension poles, or secure vehicle mounts.
- Mounting hardware is useful, and required mounting costs may be reimbursed.
How it works
Apply on OpenTrain with your resume and then complete the application on the hiring site.
About AI training work
AI training is the human work behind systems that learn from examples, including video collected and reviewed by people. Careful recording and quality checks help make those examples useful for building better AI systems.