Remote AI Trainer / AI Response Evaluation & NLP Annotation (profile-based experience)
Delivered AI response evaluation and quality rating for ML/LLM outputs, focusing on accuracy, helpfulness, and relevance to the user intent. Performed prompt engineering and testing by writing, running, and iterating prompts to improve large language model performance. Applied NLP labeling techniques such as sentiment classification and named-entity tagging to support downstream model training and dataset quality. • AI response evaluation (rating/ranking) for quality and helpfulness. • Sentiment analysis (tone/intent/emotional context classification). • Named Entity Recognition (people, places, organizations, dates). • Search relevance rating for dataset and retrieval relevance improvement.