Franck Cedric Dienzou

Evaluated, rated, and corrected the quality of responses generated by Large Language Models for international AI projects. Conducted comparative analyses of complex prompts and applied strict evaluation rubrics covering relevance, factual accuracy, tone, and safety. Drafted structured, well-reasoned justifications for each rating to improve model alignment and reliability, while processing complex information and large data volumes.

Evaluated AI model responses and scored them according to project guidelines, with special attention to borderline and ambiguous factual claims. Conducted factuality assessment and fact-checking in both French and English to reduce hallucinations and improve output reliability. Applied detailed annotation guidelines consistently across a high volume of tasks, maintaining quality standards while working efficiently.

Evaluate, rate, and correct the quality of responses generated by large language models for international AI projects. Perform comparative analyses of complex prompts and apply strict evaluation rubrics covering relevance, factual accuracy, tone, and safety. Draft well-argued, structured justifications for each rating to improve model alignment and reliability.