Expert Reviewer Contractor (AI/LLM Data Labeling and Evaluation)
As a Mercor Expert Reviewer Contractor, I contributed to data labeling and AI agent development through prompt engineering and scenario creation. I developed rubrics, identified flawed assumptions, and corrected reasoning to train and evaluate AI systems. The work focused on providing human-labeled data for professional productivity tools and auditing for a tax, accounting, and legal taxonomy project. • Designed prompts and evaluation tasks to train large language models (LLMs) in environmental and sustainability reasoning. • Authored structured scenarios and sample responses for AI model alignment. • Rated and corrected AI-generated outputs in professional productivity and legal/tax subject areas. • Collaborated with teams to refine annotation guidelines and ensure data quality.