Data Annoator
I have worked on large-scale Text-to-Image AI training projects, where I annotated and validated multimodal datasets to improve model performance and accuracy. My responsibilities included evaluating text-to-image alignment, labeling visual attributes such as objects, environments, and stylistic elements, and identifying inconsistencies or hallucinations in generated outputs. I worked with high-volume datasets, processing thousands of samples while maintaining strict daily targets and adhering to detailed annotation guidelines. Quality was a key focus we operated under structured QA systems including double-blind reviews, inter-annotator agreement checks, and internal accuracy audits to ensure consistency and dataset integrity. This experience strengthened my attention to detail, consistency under scale, and ability to follow evolving guidelines and skills that directly align with AI data labeling roles.