AI Trainer & Premium Evaluator, Domain Expert
In my role as an AI Trainer & Premium Evaluator, Domain Expert at Handshake AI, I performed response evaluation and reinforcement learning through human feedback (RLHF) for LLMs. My responsibilities included preference ranking, prompt creation, and detecting hallucinations or policy violations. I analyzed and improved model outputs with expert human feedback across complex reasoning tasks. • Performed detailed preference ranking and error flagging on LLM responses. • Designed and refined domain-specific prompts for model testing. • Provided analysis for accuracy, relevance, and policy adherence of outputs. • Contributed to LLM alignment and capability improvement in specialized domains.