AI Data Trainer & Quality Analyst at Scale AI (Remote)
Trained and evaluated large language models by labeling thousands of LLM-related data points each day to improve model performance for major tech clients. Performed sentiment analysis and entity recognition while scoring prompt response quality to reduce error rates. Collaborated with global teams to refine training guidelines and onboard new remote annotators. • Labeled 2,500+ data points daily with 99.2% accuracy • Conducted sentiment analysis, entity recognition, and prompt response scoring • Achieved top 5% annotator ranking and received a “Star Performer” award three consecutive quarters • Coordinated with teams via Slack and Zoom to update labeling practices