AI Prompt Engineer, Appen (Project: Mithrim)
As an AI Prompt Engineer at Appen, I developed AI training scenarios and evaluation rubrics to support language model evaluation. I ensured prompt accuracy, persona alignment, and compliance with project requirements throughout the data labeling process. I validated and refined submissions using the Model Mate system to maintain high-quality data for AI model performance improvement. • Designed and wrote diverse prompts according to industry and grounding standards. • Evaluated and reviewed rubric quality and prompt structure for guideline adherence. • Conducted structured validation and corrections based on system feedback. • Supported continual enhancement of AI model reliability and accuracy.