AI Training and Data Labeling Contributor
As an AI training and data labeling contributor, I worked on evaluating and improving AI-generated text responses. My tasks focused on reward modeling, ranking responses, and prompt evaluation to enhance AI alignment with human expectations. I developed strong attention to detail through the detection of bias, factual errors, and unsafe outputs. • Simplified technical explanations for clarity. • Selected higher quality answers based on clarity and usefulness. • Identified and corrected factual inaccuracies in AI-generated responses. • Rewrote biased or potentially unsafe responses to make them more neutral and inclusive.