AI Resume Analyzer: An NLP-Driven Full-Stack Platform for Intelligent Resume Screening and Candidate-Job Matching

Abstract

In today’s highly competitive job market, recruiters face an overwhelming volume of resumes that must be screened efficiently, while job seekers struggle to determine whether their resumes are tailored to specific job descriptions. This mismatch between candidate profiles and job requirements leads to inefficiencies on both sides of the hiring process. To address this challenge, the AI Resume Analyzer is proposed as a complete, web-based full-stack solution that intelligently analyses resumes against job descriptions using Natural Language Processing (NLP) techniques. The system provides candidates with actionable feedback, skill gap identification, keyword optimisation suggestions, and an overall match score, while also offering a dedicated recruiter panel for managing job postings and viewing ranked candidates, along with a super admin dashboard for overseeing the entire platform. The frontend is built using ReactJS with Vite for a fast and responsive interface, styled with Tailwind CSS, and uses React Router DOM for navigation among candidate, recruiter, and administrator screens, with Axios handling backend communication. The system implements JWT-based authentication with BCrypt password hashing to secure access across all user roles. Resume files are uploaded and stored using Multer and Cloudinary, ensuring reliable file delivery. The AI analysis engine uses pdf-parse for text extraction, TF-IDF vectorisation, and cosine similarity algorithms to compute a resume-to-job-description match score, providing candidates with precise, data-driven feedback. The platform adopts a three-tier MERN-stack architecture (MongoDB, Express.js, React, Node.js) and demonstrates how full-stack web technologies and NLP can be combined to produce an efficient, fair, and transparent recruitment tool.