AI Data Annotation & Quality Assurance
Reviewed and evaluated AI-generated responses to image-based prompts as part of an AI training and alignment project. Tasks involved assessing prompt-response pairs containing images, selecting the best or worst responses based on defined quality criteria, and providing clear written justifications for each selection. Ensured evaluations were consistent, well-reasoned, and aligned with project guidelines to help improve AI model accuracy and response quality.