Candidate Screening AI Agent: it produces the rank, a recruiter makes the call

Screening applications is where hiring tools most easily overreach. The candidate screening AI agent from Nestack Agent Care scores each application against the written requisition, keeps the ranking, the candidate notice and the impact figures together, and hands the shortlist to the named recruiter who decides who is interviewed.

What the agent does:
- Reads each application against criteria that are fixed and versioned before scoring starts, and records which criteria moved each score.
- Outputs a rank, not a rejection; where the shortlist line falls is drawn by a person.
- Records the selection rate, scoring rate and impact ratio for each category, including the unknown count.
- Holds the ranking until candidate notice is evidenced, since notice runs ten business days ahead of use under NYC's automated employment decision tool rules.
- Flags postcode, school and gaps in work history as possible proxies before they are weighed.
- Stamps every ranking with the day it ran and the adverse-impact benchmark in force that day.

Where the human stays in charge: rejecting an applicant on a score, deciding who is interviewed, setting the shortlist line, judging whether a criterion is job related and signing a bias audit all stay with people. Nobody leaves the pool except by a named recruiter, and each recruiter correction is counted in the agent's evaluation.

Read the full workflow on the Nestack candidate screening agent page, browse Nestack Agent Care, or visit Nestack.