Privacy for companies

For the people who apply to your roles, you are the controller and InterviewPilot is your processor. You decide that the interview happens, what is asked, and what the result is used for. We act on your instructions and on nothing else.

The practice side of InterviewPilot is different — there we are the controller, because the person started that session for themselves. The two are stored separately and neither can be read from the other. A candidate's practice history is never visible to you, and their interview for you never appears in their practice history.

• Their name and email address, as they gave it when applying.
• A written transcript of the interview, and any code submitted in a coding round.
• Their scores against your rubric, each one carrying the quoted answer it was drawn from.
• Session signals, where you switched them on for the role: whether the tab lost focus and for how long, pasting, leaving fullscreen, and whether more than one display is connected.
• Your decision and who made it, recorded against the application so an outcome can always be traced to a person.

Worth stating explicitly, because assessment products are routinely assumed to do all of it.

• No video and no audio recording is stored. The interview produces a transcript. If a candidate turns their camera on, the analysis runs inside their own browser and no frame is uploaded anywhere.
• No emotion or personality inference feeds your assessment. The evaluation you read is built from the transcript and nothing else. Inferring emotion from biometric data in a workplace context is prohibited outright in the EU, and we would rather not build it than build it and switch it off by region.
• No identity verification. We do not confirm a candidate is who they say they are, and no document or biometric check is performed.
• No candidate data is used to train models, yours or anyone else's, and none of it is sold or shared for advertising.

These are our sub-processors. They handle candidate data on our instructions and for no other purpose. You need this list to complete your own record of processing; it is kept accurate, and a list that omitted somebody would be a false statement rather than an oversight.

We will tell workspace owners by email before adding one, so you have the chance to object before it processes anything of yours.

Most of the list above operates from the United States, so candidate data leaves its country of origin. Transfers rely on each provider's standard contractual clauses. If your own assessment needs data to stay in one region, tell us before you run a pilot rather than after — it is not something we can arrange retrospectively.

• Interview transcripts: 180 days, then deleted automatically. Each record carries its own deletion date, so a transcript written under today's promise keeps it even if the default changes later. Export anything you need to keep.
• Feedback links sent to candidates: 30 days, after which the link stops working.
• Applications, scores and decisions: for as long as your workspace exists, or until you delete them.
• Everything, on request. Ask us to delete your workspace and the candidate data in it goes with it.

Requests about an interview someone sat for you come to you, because you are the controller. If one reaches us instead we will pass it on rather than answer it, and tell the person who to ask.

We will help you answer it: a copy of what is held, a correction, an export or a deletion. Write to support@adijain.click and we will act within 30 days, sooner where the law that applies to you is shorter.

A workspace can only ever read its own candidates. There is no cross-company view, no shared talent pool, and no administrator with a blanket read of your pipeline. Passwords are hashed. Links that work without signing in — a candidate's feedback link, email verification — are cryptographically signed so they cannot be forged or edited, and the feedback link expires.

No system is perfectly secure, and we would rather say that than imply otherwise. If we discover a breach affecting your candidates we will tell you without undue delay and with enough detail for you to make your own notification.

Before any interview begins, the candidate is told that the interviewer is an AI system, what is recorded, and whether they will receive their own feedback. That screen is generated from your role's settings, so it states a fact rather than a possibility, and it cannot be switched off.

That is a legal requirement in several of the markets you may be hiring in, and it protects you rather than us — an assessment a candidate was not told about is the one that gets challenged. What monitoring you use is your disclosure to make; that it is an AI is not optional for either of us.

A data processing agreement is available on request and we will sign one before a pilot if your process needs it. Write to support@adijain.click. The commitments in the terms for companies apply alongside it.

Material changes are dated at the top and notified to workspace owners by email. This page is governed by the laws of India.