Data Annotator
In Outlier image-to-text projects, I reviewed a wide variety of images and generated structured, descriptive text that accurately represented the visual content. The work required strong attention to detail, consistency, and the ability to identify subtle visual elements, actions, emotions, and contextual cues. I also performed quality reviews, followed project-specific guidelines, and ensured that descriptions were accurate, unbiased, and useful for training and evaluating multimodal AI systems.