Founder & Lead Engineer, AI Media Intelligence & LLM Evaluation (knew.network)
Built and managed a production-grade AI intelligence platform focused on LLM output evaluation, response adjudication, and bias annotation. Designed an evaluation pipeline that included structured scoring of authoritative, reported, and speculative outputs with tier-aware trust scoring. Conducted systematic annotation of biases across lexical, framing, omission, source selection, placement emphasis, temporal coverage, and profiling dimensions. • Evaluated and ranked LLM-generated text for factual accuracy, logical consistency, and reasoning quality. • Labeled and scored various bias factors and corroborative evidence in AI outputs. • Annotated text samples for title similarity, narrative synthesis, and trustworthiness. • Used internal/proprietary tooling for pipeline automation and structured data annotation.