Senior AI Trainer & LLM Evaluator (Freelance)
As a Senior AI Trainer & LLM Evaluator (Freelance), I contributed to RLHF preference comparison tasks and reward model training. I evaluated long-form LLM responses for helpfulness, correctness, safety, and hallucinations. I conducted red-team and safety annotation while maintaining high inter-annotator agreement. • Completed 1,000+ RLHF preference comparisons for reward model tuning • Evaluated and scored LLM outputs for quality and policy adherence • Identified jailbreaks and policy violations in safety annotation projects • Reviewed AI-generated code and reasoning for logical consistency