Annotator
The project focuses on simple, fast, and high‑volume multimodal annotation tasks, where contributors evaluate short video clips or images, describe visual content, identify differences between similar visuals, and follow clear, structured instructions to label specific features that help improve AI models; the tasks are intentionally low‑lift, highly repeatable, and designed to offer a steady flow of flexible micro‑work that supports consistent human‑guided training of visual and multimodal AI systems.