Freelance AI Trainer & Content Evaluator | Self-Employed (Remote)
Evaluated and ranked LLM-generated responses for correctness, truthfulness, helpfulness, and adherence to formatting and safety guidelines. Performed structured assessment as part of AI response evaluation workflows to identify issues such as hallucinations and rubric violations. Ensured outputs met instruction-following constraints required by each task. • Analyzed model responses against project rubrics for accuracy and compliance • Ranked responses based on helpfulness, truthfulness, and constraint adherence • Conducted deep-dive fact-checking using internet research and primary sources • Reviewed and polished English text to meet language and formatting expectations