Opeyemi Delight Johnson

AI Model Evaluator and Generalist AI Trainer with hands-on experience in LLM response evaluation, rubric-based scoring, RLHF workflows, and multimodal data annotation. I evaluate 100+ model outputs weekly, assessing accuracy, reasoning quality, and instruction adherence. Experienced in annotating text and video datasets, and supporting computer vision and robotics training through egocentric data. Background in customer service, finance and technical content writing, bringing strong research ability, linguistic precision, and clear written communication to AI evaluation workflows. Experienced working in distributed teams and adapting to evolving evaluation standards.