AI Trainer, Prompt Designer, and Reviewer
This role involved developing, refining, and testing complex prompts to train AI models, as well as evaluating model responses for accuracy, relevance, and safety. The work included creating nuanced rubrics and reviewing outputs to ensure adherence to project guidelines and data quality standards. Tasks also encompassed providing actionable feedback and improving the quality of training datasets. • Designed and iterated on prompts to enhance model capability assessment • Developed detailed rubrics for scoring AI outputs • Reviewed and improved training data quality • Delivered comprehensive feedback to other prompt writers