In my Independent Projects, I assess and rank AI-generated responses for relevance, factual accuracy, and alignment with task-specific guidelines. I also research and verify primary sources to fact-check complex claims.
I test model boundaries for unsafe hallucinations, partial compliance, and delayed compliance, and annotate text and images for sentiment, emotion, and user intent. I flag ambiguous edge cases and provide evidence-based justifications.
My physiotherapy training informs my clinical healthcare knowledge and fact-checking. During supervised placements and volunteer work, I assisted with patient assessment and treatment and maintained confidential clinical records.
