Independent Data Annotator & Content Specialist (LLM content evaluation and dataset label review)
TextIndependently evaluated and graded machine-generated text outputs from experimental LLMs for truthfulness, logic, safety, and compliance with complex quality rubrics. Performed human-in-the-loop style review by identifying errors and checking adherence to strict user-defined negative constraints. Conducted independent research to verify factual integrity and detect hallucinations and logical fallacies.• Evaluated model performance against multi-criteria quality rubrics and negative constraints.• Classified and reviewed dataset labels to improve conversational accuracy and model behavior.• Produced precise analytical explanations of findings for downstream stakeholders.• Verified claims using primary and credible secondary web sources to maintain data integrity.