AI Trainer – Biology Domain (micro1 & Meridial/Invisible, Remote)
Evaluated AI-generated biology and healthcare responses for factual accuracy, reasoning quality, and scientific validity. Performed rubric-based ranking and assessment of LLM outputs to determine alignment with biomedical expectations. Identified hallucinations and scientific inaccuracies and provided structured feedback to improve response quality. • Assessed LLM responses against biomedical factuality criteria • Used rubrics to rate and rank outputs • Performed hallucination detection for biomedical content • Delivered feedback to support AI alignment and quality improvements