HireBot
Towards Automation of the Hiring Process
Created on 30th June 2020
•
HireBot
Towards Automation of the Hiring Process
The problem HireBot solves
HireBot is built to automate the hiring process of the industry. It bridges the gap between the candidates and HRs of companies. Most of the times candidates are rejected without proper screening as there are a lot of applications and it is not possible for a human to go through all of them properly. Here comes in the HireBot, companies can just provide the link to the test for the required profile having personalised questions and candidate records the video answers, which are then analysed by Deep Learning on our servers and identify different traits such as condifence, happiness, engagement, etc. and a score is generated based on the content of answers and traits.This all information can be viewed from company Dashboard. Also a threshold score can be set to screen candidates automatically for further in person interviews.
Challenges I ran into
The biggest challenge was how to analyse confidence by video. As we did find any accurate and fast enough simple algorithm, we came up with our own. We tracked the face and eye movement of the candidate for a small time frame, took the average position and find out how much the face and eye are diverting from their average position, Based on this a score is assigned, too less movement generally shows less interaction and interest, while too much movement is also not good. We used OpenCV for this video analysis task.
Technologies used