A Modern Job Portal with Built-in AI Mock Interviews

Authors:

Prince Kumar Giri, Dharjya Jalota, Ravivardhan

Abstract

As a result, there is a need to build a web application that would allow integration of job hunters, employers and administrators of the platform in a single platform. The paper presents the design and development of the web application of the Job portal using the MERN stack (MongoDB, Express.js, React.js and Node.js) with emphasis on the three types of users - Admin, Job Hunter, and Recruiter.

The Job Hunter type of users will be able to create a detailed profile, search for jobs, apply to jobs and track applications. The functionality of the recruiter type of user includes opening the company account, publishing the job vacancies, managing vacancies and reviewing applicants from job hunters. Finally, the admin portal presents information provided by the system and shows the results using Recharts charts.

All security control measures are applied throughout the system using JSON Web Token as well as role-based access control so that all user types will only be permitted to do what they are allowed to. In terms of security controls, during the authentication process the method of hashing passwords is bcrypt and request input data is validated by the Express Validator before proceeding to the business logic. In terms of the front-end architecture, React.js with Tailwind CSS and React Query has been used to achieve efficient and highly efficient caching mechanism for the user interface while in the back-end architecture Model-View-Controller architecture using REST has been applied.

The feature that is considered as the best service and makes this platform stand out from the crowd of other recruitment platforms is the Mock Interview Practice that enables all users to practice the technical competency on the relevant questions and get immediate feedback on their performance. The deployment strategy for this application is done via Vercel and the database cloud used is MongoDB Atlas with a very good API latency (less than 500ms) and Google Lighthouse (more than 85%).