The service helps job candidates do their preliminary interview as successfully as possible. To get the service, the user needs to download a video file with self-presentation where he will tell what he’s going to say during the interview.

Then the neural network from Russian startup VisionLabs processes the files and gives its evaluation on the basis of the poll of 537 HR specialists from different industries of business. According to the developers, the neural network has already been trained on more than 1000 video files with self-presentations and learned how to calculate the probability of a successful interview.

The expert proceeds to view the record and points out the mistakes in the presentation. The neural network and expert evaluate what the candidate is saying, his intonation, speech speed and loudness, as well as gestures, pose and mimics. As a result, the user receives a written report (example) on the probability of successfully passing the interview and the results on every evaluated factor.