AI Data Annotator & Prompt Response Evaluator
As an AI Data Annotator & Prompt Response Evaluator for the Aurora Project at Lionbridge AI, I evaluated prompts and AI-generated responses for accuracy, safety, and usefulness according to project guidelines. I annotated and labelled text data, reviewed language content, and identified errors to maintain high-quality datasets. Tasks included ranking prompt completions, validating written outputs against audio, and supporting computer vision datasets via recording and compliance checks. • Maintained a 96–98% quality acceptance rate across prompt evaluation and annotation tasks. • Performed transcription and validation by cross-checking audio and text data for completeness and accuracy. • Conducted voice, face, and video recording tasks for computer vision datasets. • Adapted quickly to updated annotation guidelines and flagged edge cases to improve data quality.