AI-Powered Resume Evaluation System

Abstract: The contemporary job market is highly competitive, with Applicant Tracking Systems (ATS) functioning as the main tool for evaluating resumes.

Although these systems are efficient, they often depend on strict keyword matching, which may not fully capture the meaning behind a candidate's experience. This paper introduces an intelligent framework intended to overcome this limitation by using the advanced natural language processing and generation abilities of Large Language Models (LLMs). Our system goes beyond basic resume parsing by extracting essential information from PDF resumes and offering a range of career improvement tools. These include a detailed ATS compatibility score with specific recommendations, recognition of skills that are currently popular and relevant to a candidate's background and desired positions, and tailored suggestions for professional certifications. By making use of the Mistral-7B-Instruct-v0.2 model through the Hugging Face Inference API, our framework offers a practical, easy-to-use, and effective method to support job seekers with insights that are backed by data, helping them make informed decisions about their careers.