AI Data Trainer — Scale AI
Reviewed and labeled AI-generated text responses to support training of large language models. Conducted quality assurance checks to ensure annotation consistency and high task accuracy. Assisted with evaluation of multilingual transcription datasets and conversational AI outputs. • Labeled AI responses for LLM training workflows • Performed QA accuracy reviews (98% task accuracy) • Collaborated remotely with AI researchers and project managers • Supported multilingual dataset evaluation and conversational output assessment