In my current independent AI evaluation work, I assess text, image, and conversational outputs for accuracy, relevance, instruction following, naturalness, task success, and consistency.
I compare model responses, look for error patterns, and write concise, evidence-based rationales. At Remotasks, I worked on text, image, and audio annotation and guideline-based quality control; at UHRS, I rated search relevance, text quality, and content usefulness.
My public health experience includes supporting nutrition services, counseling, monitoring, and program reporting at Puskesmas Pa'bentengang, Puskesmas Bahaur Hilir, and Puskesmas Luas.
Earlier, I conducted structured field data collection for the UNICEF WinS Survey and supported research on elderly caregiving practices. I also worked in sales coordination and distribution-center administration.
