AI Trainer (Freelance)
Evaluated and ranked AI model responses based on helpfulness, relevance, accuracy, and safety following detailed instructions and rubric criteria. Labeled and annotated text datasets for tasks such as text classification, intent detection, and content moderation to support downstream model training and evaluation. Ensured outputs were neutral, consistent, and aligned with provided guidelines to maintain quality and productivity targets. • Performed response ranking and quality scoring across multiple evaluation dimensions • Labeled intent and categories for text classification tasks • Moderated content according to safety and guideline rules • Met productivity expectations while maintaining annotation accuracy