Recruit AI: An AI-powered intelligent recruitment management system
Department of Information Technology, Sanjivani Pratisthan Institute of Technology (S.P.I.T.) Polytechnic, Kurund, Tal-Parner, Dist. – Ahilyanagar, India.
Research Article
International Journal of Engineering Research Updates, 2026, 09(01), 009-020.
Article DOI: 10.53430/ijeru.2026.9.1.0015
Publication history:
Received on 12 February 2026; revised on 18 March 2026; accepted on 20 March 2026
Abstract:
The recruitment process in modern organizations faces significant challenges including high volume of applications, time-consuming manual screening, and subjective candidate evaluation. This paper presents Recruit AI, an intelligent web-based recruitment management system that leverages artificial intelligence and machine learning to automate and optimize the hiring workflow. The system implements a hybrid resume matching algorithm combining TF-IDF vectorization and cosine similarity to achieve accurate candidate-job matching with scores ranging from 0–100%. Additionally, it integrates Samba Nova’s Meta-Llama-3.1-8B-Instruct model for conversational AI assistance and employs automated email workflows using Django Signals. Experimental results demonstrate that the AI-powered matching algorithm achieves an average accuracy of 85% in identifying suitable candidates, reducing manual screening time by approximately 70%. The system serves dual user roles (candidates and recruiters) through a responsive web interface built with Django 5.x framework, featuring profile management, application tracking, and comprehensive analytics dashboards. This research contributes to the field of HR technology by presenting a scalable, cost-effective solution that enhances recruitment efficiency while maintaining transparency and fairness in candidate evaluation.
Keywords:
Recruitment Management; Artificial Intelligence; Resume Matching; Tf-Idf; Cosine Similarity; Natural Language Processing; Django Framework; Machine Learning
Full text article in PDF:
Copyright information:
Copyright © 2026 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0