About the Role
Help teach the next generation of AI how to understand the world. We are looking for Description Annotators who can transform verified video clips into clear, detailed, and structured natural language descriptions that power cutting-edge Vision-Language-Action AI models.
What you will do
Write accurate, object-focused descriptions of actions in short video clips
Independently describe the actions performed by each hand when both hands are visible
Label camera movement (Small or Large)
Select the appropriate scene from a structured vocabulary
Ensure every description is precise, consistent, and easy for AI models to learn from
What makes a great description?
Instead of writing "Washes the bowl", you will write "Washes the blue and white ceramic bowl with a blue sponge near the faucet."
What we look for
Excellent written English
Strong attention to detail
The ability to observe and describe actions accurately
A laptop and reliable internet connection
Availability to work remotely
Note: No prior annotation experience? We provide training for motivated candidates with strong observation and writing skills.