Lead AI Trainer & Technical Data Annotator (Remote Contract)
Performed evaluation, ranking, and rewriting of high-complexity multi-turn LLM responses using strict quality dimensions. Created ground-truth datasets to optimize developer prompts for function calling and API integration. Reviewed automated annotation workflows, updated labeling taxonomies, and validated dataset accuracy using edge-case discrepancy checks. • Truthfulness, helpfulness, logic, and constraint adherence grading • Ground truth dataset authoring for API/function-calling developer tasks • Edge-case identification and taxonomy updates to improve accuracy • QA of automated annotation workflows and dataset validation