Co-authored a peer-reviewed study examining whether researchers can reliably identify the same corrective-feedback episodes in naturalistic online tutoring. The work foregrounds coding transparency, inter-rater reliability and audit trails before substantive claims are made.