AWS-Based Automated Applicant Tracking System for Job Matching and Recommendation
Abstract: The rapid growth of online recruitment platforms has created a large volume of job postings, making it difficult for candidates to identify positions that closely match their skills and professional experience. This paper presents an AWS-based automated Applicant Tracking System (ATS) for job collection, processing, candidate-job matching, recommendation, and notification. The proposed system collects job postings from multiple sources and stores raw data in Amazon Simple Storage Service (Amazon S3). AWS Lambda is used for serverless ingestion and orchestration, while AWS Glue performs extraction, transformation, normalization, and conversion of job data into a query-efficient processed format. Amazon Athena provides SQL-based access to the processed data. Two matching approaches are implemented: a traditional ATS approach using textual similarity, skills, job title, and experience information, and a machine-learning-based approach combining TF-IDF, BM25, Sentence-BERT semantic similarity, skill matching, and experience matching. Selected recommendations can be delivered through Amazon Simple Email Service (Amazon SES). The architecture is designed to be scalable, modular, and suitable for automated job recommendation workflows. Experimental evaluation includes threshold sensitivity and matching-score analysis to study the behavior of the recommendation pipeline.
Keywords: Applicant Tracking System; AWS; BM25; Job Recommendation; Machine Learning; Sentence-BERT
How to Cite?: Bammidi Lokesh, "AWS-Based Automated Applicant Tracking System for Job Matching and Recommendation", Volume 15 Issue 10, October 2026, International Journal of Science and Research (IJSR), Pages: 269-272, https://www.ijsr.net/getabstract.php?paperid=SR261003132556, DOI: https://dx.doi.org/10.21275/SR261003132556