Senior AI Data Labeler & LLM Trainer
Managed and executed over 15,000 high-quality annotated items for leading AI platforms, primarily using RLHF methods. Focused on Chinese-English preference ranking, instruction following, text evaluation, and red teaming. Maintained a 97%+ accuracy rate while exceeding daily quotas by 40%. • RLHF preference ranking and response evaluation • Daily output up to 800 items • Chinese-English bilingual annotation • Used LabelStudio, Scale AI, Appen, Doccano, CVAT, Excel, Google Sheets