Independent AI Trainer / Evaluator / Creative Tester (AI dialogue & narrative evaluation)
Conducted structured evaluation and annotation of AI-generated dialogue, narratives, and character interactions using internal rubrics. Designed test prompts and scenarios to score tone, coherence, and world consistency across iterative model outputs. Built behavioral and dialogue-evaluation frameworks to preserve character continuity and consistency over time. • Evaluated outputs against quality and emotional-depth criteria • Maintained evaluation datasets for ongoing model improvement • Authored structured documentation of evaluation workflows • Tuned local model configurations to support multimodal/creative testing