Builds the screen from the gaps: an unverified must-have becomes a question. Answers are saved as the candidate's own claims, kept apart from documented evidence.
The call covers what's already on paper, and the missing must-have comes up at the client interview instead. Questions tied to the job, asked the same way each time, predict performance better than an open chat.
.42 vs .19
mean validity of structured and unstructured interviews for predicting job performance.
Each unverified or partial must-have becomes a question, drawn from your industry pack's question bank.
The questions that can end the process come first, so a short call can stop early.
Answers are recorded as the candidate's own words, kept apart from documented evidence.
Request: Build the screen for Dana R. on the ASR III req.
Request: Build the screen for Priya S. on the desktop support req.
A preview with fictional data. Screening Questions is coming soon.
Screening Questions is coming soon. The approval rules below already apply to every agent in ANDY today.
Changes to clearance, salary or work authorization that come out of a screen.
Each answer, who recorded it and when, kept as a claim until a document backs it.
Every requisition shows its mode: Co-Pilot, Hybrid or Custom. Co-Pilot, the default, sends nothing without your approval.
How approval works