AI Data Trainer at DataForce by TransPerfect (LLM evaluation and dataset annotation)
Conducted LLM-related review and scoring of AI-generated outputs for relevance, tone, and factual accuracy. Supported dataset quality by verifying consistency with labeling guidelines. • Assessed responses for coherence, relevance, and factual correctness • Labeled and classified NLP and computer-vision datasets • Partnered with QA to improve labeling guidelines and consistency • Performed iterative checks to maintain data integrity