Business Partner & AI Prompt Evaluator (Self-employed, Source Corporate)
Designed, authored, and engineered accurate AI prompts for evaluating model responses and improving data quality using state-of-the-art LLMs. Developed python-based codes and prompt strategies focused on AI-driven data quality improvement, duplicate detection, and result evaluation. Evaluated outputs from models such as GPT 5.4, Claude Opus, Claude Sonnet, and Google Gemini in various engineering and business use cases.• Built and fine-tuned prompts and responses for LLM evaluation and supervised AI model behavior. • Oversaw evaluation, rating, and refinement of AI-generated output from several platforms. • Collaborated with engineering and business teams to construct high-quality data labeling pipelines. • Contributed to improvements in data normalization, classification, and de-duplication for diverse ERP sources.