用 Jev 从通话日志预测预约成败
只喂 AI 接待员的结构化事件日志（身份校验失败、查询空档、来电沉默），每轮让 Jev 预测这通电话会不会成交，再与真实病历系统结果比对。
Every AI receptionist call leaves traces, like 'identity check failed', 'searched for open slots', 'caller went silent', with timestamps. We fed only those logs, no words, to Jev. After every agent turn we asked it 'will this call end in a booking?', then compared its guesses to what really happened in the medical record system.
We gave Jev 2,029 real phone calls. No transcripts or audio; it never heard a word. Our AI receptionist's calls were reduced to pure structure, meaning turns, tool calls, workflow stages and timing. During the calls, Jev made 38,012 turn-level forecasts at 118 ms median latency, reviewed every call with five typed questions and produced 10,145 answers in 26 seconds with 256 requests in flight. The experiment was zero-shot, with no fine-tuning or examples from our data. We compared Jev's forecasts with what actually happened in the EHR. By the halfway point, Jev could meaningfully separate calls that would book from those that wouldn't (AUC 0.78), and near the end it ranked them correctly 94% of the time. Even though Jev over-focused on visible errors our agent usually overcomes, it's still pretty incredible that it analyzed thousands of real calls in seconds for only $3.