Wesley Ekpenyong

about

I help teams build more reliable AI systems through careful data annotation, response evaluation, and quality assurance. My experience includes labeling text, images, audio, and video; evaluating search relevance; moderating content; and testing chatbot conversations for logical consistency. I review AI-generated responses for accuracy, relevance, and safety, cross-check factual claims, and document hallucinations, bias, and reasoning errors with actionable feedback. I have worked with Atlas Capture, OpenTrain AI, TELUS International AI, and RWS Train AI on AI training and data quality projects. I’m comfortable validating annotated datasets, resolving labeling discrepancies, and collaborating with remote teams to refine guidelines and adapt to changing requirements. I bring particular attention to culturally sensitive content and inclusive annotation practices. With certifications in Azure Data Science and IBM Data Science, I also bring a foundation in data analysis, statistical modeling, and machine learning. I can support projects that need accurate labels, thorough quality reviews, and clear feedback to improve training data and model outputs.