AI Trainer (Contract) | Handshake AI Fellowship / AI Training Platform
As an AI Trainer, I was responsible for evaluating AI-generated text and images for accuracy and relevance. I performed structured ranking, labeling, and categorization of multimodal datasets, focusing on identifying hallucinations and logical inconsistencies. I executed tasks including model output comparison, prompt-response analysis, and adversarial evaluation techniques. • Conducted evaluation of both text and image outputs for accuracy and logical consistency • Labeled and categorized data to support AI model training and validation • Applied adversarial prompt techniques to reveal model weaknesses • Performed structured model comparisons and provided feedback on model performance