Hiring interviews: debrief on evidence, not memory
Interview debriefs run on the interviewer’s memory, which is to say on first impressions wearing a lab coat. What did the candidate actually say about the outage — and was it the same story they told the other panel? Donna keeps the verbatim record and reads each session closely, so the debrief argues about evidence. And because candidate interviews are personal data with legal retention rules, they belong on your server, under your policy — not in a notetaker vendor’s indefinite cloud.
What goes wrong without a real record
- Debriefs reward confident recall over accurate recall — and they are rarely the same interviewer.
- Candidates tell different panels different stories, and nobody cross-checks until the reference call.
- Candidate recordings on third-party clouds collide with data-protection law and your own privacy policy.
What Donna catches here
The candidate’s actual claims — team size, ownership, results — verbatim and timestamped, ready to verify in references instead of half-remembered in the debrief.
The question about why they left that got answered with a story about why they joined. Dodges are signal, and Donna logs every one.
Where the candidate revised their own account under a follow-up — the difference between polish and consistency, visible on the page.
Panel dynamics too: which interviewer talked more than the candidate, whose questions got cut off. Interview quality is also being interviewed.
All eight sections of the report, explained: meeting intelligence vs meeting notes.
Questions, answered straight
Is recording interviews with an AI bot fair to candidates?
With disclosure and consent, it often makes the process fairer: decisions get grounded in what the candidate actually said rather than the loudest memory in the debrief. Donna joins visibly, so disclosure is structural, and self-hosting lets you honor your stated retention policy with a delete on your own disk.
Does Donna help with structured interviews?
Yes — the report makes deviations visible: questions skipped, answers that wandered off-rubric, and inconsistencies between what the same candidate told different panels across sessions stored in your dashboard.
How long are interview recordings kept?
As long as you decide and not a day more — files sit on your server’s disk and rows in your PostgreSQL. Retention is a policy you enforce with your own cron job, not a vendor setting buried in an enterprise tier.
Put Donna in your next meeting
Donna deploys onto your own VPS in an afternoon: nginx, pm2, PostgreSQL, your API keys. Early access is open — tell us about your team and we’ll get her a seat at your table.
Request early access