Inspiration
The first round of interview is mostly coding where we are provided problems on HackerRank or some other platform. Now the trick that many people use it, they ask some of their friends to do it on their behalf and move easily to the next round. The whole process of hiring is very tiring and involves many useful resources and money. So if cheat can be detected at an early stage, it will have multiple benefits. Major of these are:
- Cutting down possibilities of Cheating
- Only good candidates are passed through
- Saves time and resources of interviewers
What it does
We are using several events to check if the candidate is doing cheating such as:
- Screensharing through Chrome Desktop share, Teamviewer and Skype
- If multiple people are present for the challenges
- Using Print screen
How we built it
We have noticed some trend over screen sharing, such as when using Google Chrome it shows text at the top such as "Your desktop is currently shared with jayrodge15@gmail.com. Stop Button(Button)".
Challenges we ran into
Using Google Cloud Vision API, at first we thought it would be easy, but it turned out to be quite difficult because input type that it was expecting and the input type of our screenshots was different.
Accomplishments that we're proud of
We are successful in detecting whether the candidates are using screen share when the coding challenge is active and whether there are more than one people.
What we learned
Calling APIs isn't easy, figuring out whats APIs expect and using it for the application takes time.
What's next for Online Coding Challenges Cheat Detection
Using Computer Vision and Deep Learning to detect various screen sharing tools such as Skype or any other third-party applications.

Log in or sign up for Devpost to join the conversation.