An LLM-Based Resume Intelligence System for Recruiters and Job Seekers
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
The increasing volume of job applications in modern recruitment workflows has revealed clear limitations in traditional resume screening methods, which largely depend on keyword matching and basic filtering techniques. These approaches often struggle to capture contextual skills, lack deeper semantic understanding, and provide little to no feedback for candidates. Additionally, most existing systems are designed primarily for recruiter-side filtering, offering minimal assistance to job seekers. This paper presents an LLM-Based Resume Intelligence System that introduces a dual-mode framework to support both recruiters and job seekers within a unified architecture. The system employs Large Language Models to perform context-aware resume parsing, structured information extraction, and semantic skill interpretation. To enable scalable similarity analysis and efficient retrieval, resumes and job descriptions are transformed into vector embeddings and indexed using FAISS. In recruiter mode, the system facilitates automated bulk resume screening, role-fit evaluation, and candidate ranking based on semantic relevance. In job seeker mode, it provides resume insights, identifies potential skill gaps, and generates targeted improvement guidance. The proposed design emphasizes scalability, explainability, and practical deployment feasibility by avoiding computationally expensive multi-model ensembles. Preliminary evaluations using ranking-oriented metrics, including Precision@K and Mean Reciprocal Rank (MRR), indicate consistent retrieval behaviour and reliable role-fit assessment. The results suggest that integrating LLM-driven semantic analysis with vector search techniques can enhance recruitment efficiency while providing meaningful decision support for both stakeholders.
Fields of Study & Subjects
AI and HR Technologies
Recommender Systems and Techniques
Information Retrieval and Search Behavior
Publication details
- Journal / source
- Research Digest on Engineering Management and Social Innovations
- Publication date
- 31 March 2026
- Language
- EN