Tool

Verdict — Adversarial AI debate.

Two AI advocates argue for each candidate citing evidence. Independent judge evaluates with anonymized candidates. Evidence-based decision, not gut feel.

Open Verdict

Need role calibration, candidate evidence map, and post-hire tracking too?

Verdict is included in Talent Validation, our full hiring decisions cockpit.

See Talent Validation →

Two modes

Depending on your candidate pool.

1v1 mode

Compare two candidates directly. 3 debate rounds. Output: scored verdict + PDF report.

Swiss Tournament mode

Rank up to 32 candidates in 4-5 rounds. Top 4 go to playoffs. Output: ZIP with individual reports.

Why adversarial debate works

Two AI advocates with access to the same CV build opposing arguments. An independent judge evaluates without knowing who argued what. The format forces evidence to be made explicit — and exposes the gaps that gut feel hides.