Two-Agent Resume Screening Increases Application Pass Rates via Evidence Exchange

A bilateral two-agent framework using GPT-5.5 and Claude Opus 4.7 improves resume screening by allowing employer and candidate agents to exchange evidence. In borderline cases, the pass-instance rate rose from 4.5% to 26.2% for GPT-5.5, though it also changes existing one-call decisions.