AI Trainer (RLHF and Model Evaluation)
I participated in reinforcement learning from human feedback (RLHF) to enhance the performance of large language models. My role involved evaluating, ranking, and providing nuanced feedback on AI-generated outputs to ensure accuracy, safety, and linguistic quality. I focused on refining LLM behavior using advanced linguistic strategies and bias detection techniques. • Provided text-based evaluations and rankings for a variety of language understanding tasks. • Applied cognitive psychology principles to improve prompt engineering and output validation. • Collaborated with AI tool developers using platforms such as ChatGPT-4, Claude 3, and Google Gemini. • Developed and iterated on diverse prompt templates to optimize model alignment with human communication.