ARCHIVES

The recruitment process in many organizations continues to rely heavily on manual resume screening, making it a time-consuming, labor-intensive, and error-prone activity. Human involvement in initial screening stages often introduces unconscious bias and inconsistency, which can negatively impact fairness and the overall quality of candidate selection. To overcome these challenges, this paper presents an AI-powered automated resume screening system that utilizes the Google Gemini 1.5 Pro API to enable deep contextual understanding and intelligent evaluation of candidates. The first step in the suggested system is to extract text from resumes in various formats, including DOCX and PDF. After that, advanced Natural Language Processing (NLP) techniques are used to find pertinent keywords, credentials, professional experience, educational background and skills. Several evaluation metrics, such as an Applicant Tracking System (ATS) score and a detailed skill match percentage, are calculated by comparing the extracted data with job descriptions. Additionally, the system helps candidates improve their resumes to better fit particular job roles by offering helpful and customized improvement suggestions. Within a micro service-based architecture, Flask is used for backend services, MongoDB is used for flexible and safe data storage, and Next.js is used for a responsive user interface. Scalability, modularity, and safe communication between services are guaranteed by this design. The suggested solution drastically cuts down on hiring time, increases evaluation accuracy and consistency, and fosters fairness by reducing human bias by automating resume screening. All things considered; the system improves recruiters' decision-making effectiveness while providing job applicants with insightful feedback.