Freelance AI Data Annotator & Model Evaluator (Remote)
Freelance AI data annotation and model-output evaluation across multiple AI training platforms, focusing on scoring and ranking responses using structured rubrics. The work included reviewing for accuracy, coherence, helpfulness, and factuality, as well as identifying hallucinations, logical errors, and code defects. Tasks also involved prompt/response critique and independent quality assurance across large task batches. • Labeled and evaluated chat-response quality and general-knowledge answers • Applied rubric-based rating and preference rankings between model outputs • Conducted red-team style checks for hallucinations, inconsistencies, and reasoning issues • Reviewed and corrected Python/SQL snippets for correctness, efficiency, and style