Request a free trial

Take a 2-minute tour of how Codility handles cheating in technical hiring

Worried that candidates cheat on technical assessments with AI or someone else’s help? Follow one backend assessment from setup to the report, where every flag comes with evidence a person can review.

What you’ll see in the tour

8 steps through one backend engineering assessment: Bianca sets the integrity rules, Jonah takes a task with AI off, and Bianca reviews the evidence in the report.

  1. 01

    Can candidates use AI where you don’t allow it?

    Choose AI on or off for each task. Where it’s on, every prompt and answer is captured as reviewable AI activity. Where it’s off, the task runs with no AI Copilot or Claude Code.

  2. 02

    What about hidden AI tools that sit over the candidate’s screen?

    Cheating apps detection runs in the optional Codility desktop app. It’s designed to surface hidden AI tools and anyone connecting to the computer remotely, and it keeps the session in full screen until the candidate submits.

  3. 03

    Do candidates have to install anything?

    Only when you turn on Cheating apps detection for an assessment. Candidates then download the Codility desktop app, and they’re told before they start.

  4. 04

    What if someone else takes the test, or reads answers off a second screen?

    ID verification checks a photo ID before the candidate starts, and the network check shows whether the candidate stayed on one network. The typing pattern signal looks for code retyped from a second screen or another device.

  5. 05

    What if your questions leak online?

    Video task statements narrate and animate the problem, so there is no written statement to copy or post. The similarity check compares each solution with other candidates’ work and solutions found online.

  6. 06

    Who decides?

    A person. Signals roll up into one integrity risk rating from None to High, worked out the same way every time, and replay shows the evidence behind a flag. No automated hiring decisions are made by AI or machine learning.

Built on almost two decades of technical assessment

1,300+validated tasks in 80+ languages and frameworks
9 in 10candidates say the assessment content is fair, from 45,925 survey responses
4.6out of 5 on G2, from 900+ reviews
  • SOC 2 Type II
  • ISO 27001
  • GDPR
  • CCPA
Visit the Trust Center

See how Codility handles cheating in your own hiring

Tell us which engineering roles you hire for and what worries you most about assessment integrity, and we’ll walk you through the settings, the AI policy and the report review that fit your process.

Talk to us