Multimodal AI Data Trainer
ImageThis role involved evaluating and annotating AI-generated outputs across text, image, audio, and video datasets to support model training and refinement. I performed structured data labeling, prompt analysis, response ranking, and provided detailed human feedback to enhance AI model quality. Quality assurance, accuracy verification, bias detection, and strict adherence to project guidelines were critical components of my daily workflow. • Evaluated and refined model outputs for accuracy, factuality, safety, and coherence. • Performed pairwise comparisons, prompt evaluation, and response ranking in multimodal AI pipelines. • Detected hallucinations, inconsistencies, and bias in AI-generated content for quality improvement. • Collaborated remotely on scalable AI training, annotation, and evaluation projects with high accuracy standards.