LLM Evaluation AI Automation Specialist & Systems Architect
Provided an end-to-end workflow for LLM evaluation and quality-oriented prompt iteration aligned with RLHF and LLM evaluation practices. Built systems that generate, test, and benchmark outputs for downstream automation. Supported improving model response usefulness for business clients. • Documented and operationalized prompt/workflow variations for evaluation • Evaluated outputs across quality dimensions (accuracy, tone, task completion) • Integrated evaluation results into automated reporting and notification pipelines • Used AI workflow design to reduce manual effort in client operations