Generalist Freelance AI Model Evaluator (Confidential AI Training Platform)
Text
Prompt + Response Writing (SFT)Created and authored high-quality multi-turn prompts and detailed grading rationales to support evaluation and training of LLM outputs. Developed rubrics and structured guidance to ensure consistent scoring across generative AI responses. Produced training data artifacts suitable for iterative model improvement and continuous quality checks. • Wrote multi-turn prompts for LLM interactions • Authored grading rationales in Markdown format • Designed rubrics to standardize evaluation • Performed ongoing prompt evaluation and labeling for model training
2026 - Present