Freelance AI Data Specialist & Content Evaluator (Various AI Platforms: Outlier, DataAnnotation, etc.)
Freelance AI Data Specialist & Content Evaluator performing LLM response evaluation and ranking using strict rubrics. Tasks included truthfulness, helpfulness, and safety judgments, along with RLHF activities to improve model behavior and reduce hallucinations. Work also involved identifying and logging edge cases to refine platform-specific training guidelines. • Evaluated LLM-generated outputs for accuracy and compliance with guidelines • Ranked complex responses using structured rubric criteria with written justifications • Conducted RLHF tasks focused on improving logic and reducing hallucinations • Logged edge cases for dataset and guideline refinement