Today, in recruitment, the volume of applications makes the traditional resume screening ineffective, discriminatory, and time-consuming. This paper presents a solution based on AI that combines Natural Language Processing (NLP) and Machine Learning (ML) to optimize and automate the resume shortlisting process. The model incorporates OCR, feature extraction, and AI-based ranking to grade candidates in terms of skills and job fit. Comparison with models such as Random Forest, SVM, Logistic Regression, LSTM, and BERT validates BERT’s superiority in both accuracy (90%) and relevance score (88%). Despite the longer time required for processing, it is BERT’s contextual understanding and broader language representation which enable it to overpower other models. The system not only eliminates human error but also augments the recruitment process and ensures equal treatment in hiring. The next steps could involve improving acceleration and using it for different industries. This study has contributed to AI-based recruitment, hiring intelligence and data-intensive.