AI cheating in remote interviews is now the norm, not the exception.
Candidates are running Cluely, Parakeet AI, and screen
overlay tools while sharing their screen. The interviewer sees clean code
being written. The candidate is getting it fed to them in real time.
The worst part? There's no way to tell. Not from the video. Not from the
screen share. Not from how confidently they answer follow-up questions —
because the AI answers those too.
We started talking to engineering managers and technical recruiters. The
stories were consistent:
The hiring signal from remote technical interviews has become unreliable.
Everyone knows it. Nobody had a good solution.
Zero Assist is a desktop app candidates install before the interview starts.
They see exactly what it monitors — full transparency, explicit consent.
Then the interview begins and the app runs quietly in the background.
No pop-ups. No interruptions. The candidate experiences a completely
normal interview.
The interviewer, on the other hand, gets a live dashboard showing:
If something is detected — a ChatGPT window, an AI coding assistant, a screen overlay tool — the interviewer sees it immediately and can decide
how to respond. No confrontation required mid-interview.
We specifically didn't build a covert monitoring tool.
Candidates know Zero Assist is running. They see the pre-flight check
before the session starts. This does two things: it deters most cheating
attempts before they even start, and it makes the data legally defensible
for companies that need documented hiring records. The goal isn't to catch people. It's to restore the integrity signal that
remote interviews have lost.
Zero Assist integrates with Greenhouse, Lever, and Ashby — so sending
the monitoring link to candidates fits directly into existing ATS workflows.
No new tools for recruiters to learn.
If you're running remote technical interviews and the signal feels broken, this is what we built to fix it.