Persian (Farsi) AI Trainer & AI Rater — Invisible Technologies (Freelance)
Evaluated and rated AI-generated Persian outputs for accuracy, fluency, cultural appropriateness, and instruction adherence to support LLM improvement. Applied biomedical domain expertise to assess task-specific correctness and provide structured feedback. Delivered localization-oriented guidance to improve safety and quality of domain-specific model responses. • Rated AI outputs against multilingual quality and instruction-following criteria • Reviewed biomedical factual accuracy and domain consistency • Provided structured feedback for model refinement and localization • Ensured outputs align with safer, domain-appropriate response expectations