Task Auditor
As a task auditor for AI training data, I performed high-level quality control, acting as the final verifier to ensure the integrity of labeled data and AI-generated outputs. I prioritized accuracy, strict guideline adherence, and unbiased data validation utilizing advanced labeling taxonomies. This position demanded meticulous preference ranking and structured feedback to align model performance with project expectations. • Conducted in-depth quality assurance reviews of AI and human-labeled outputs. • Enforced inter-annotator agreement, ensuring consistency across annotators according to rubrics. • Identified and corrected data biases for fair and representative AI model training. • Provided critical feedback to engineering teams to refine labeling standards and project guidelines.