faith Nzeli

I evaluate AI-generated responses for accuracy, relevance, clarity, and instruction compliance, using structured feedback to improve model performance and reliability.

I've supported AI training through data review, annotation, validation, response ranking, and dataset quality checks. My work spans text, image, and multimedia outputs, with close attention to errors, inconsistencies, safety, and usability.

I develop and test prompts to assess AI reasoning, factual accuracy, structure, and user intent. I refine prompt effectiveness through testing and apply detailed evaluation criteria to large-scale AI evaluation tasks.

With a background in Computer Science and Artificial Intelligence from Carnegie Mellon University, I bring knowledge of machine learning, NLP, data analysis, quality assurance, and AI training workflows.