CV Sorting System

Our project aimed to streamline and enhance the recruitment process by developing a CV sorting system powered by Agentic AI, leveraging the capabilities of a Large Language Model (LLM). The system was designed to intelligently evaluate, classify, and rank resumes based on job-specific requirements, skills, and experience, significantly reducing manual effort in the initial screening phase. Using Agentic AI architecture, the system demonstrated autonomy in understanding job descriptions, extracting key qualifications, and matching them with the relevant content in each CV. Unlike traditional keyword-matching algorithms, our model utilized the contextual understanding of LLMs (e.g., GPT-based models) to assess the overall relevance and coherence of each candidate profile. This enabled it to interpret nuanced information, such as soft skills, project involvement, and role-specific experience. The system was trained and tested on a diverse dataset of anonymized resumes and job postings. Performance metrics showed a significant improvement in accuracy and time efficiency compared to conventional methods. It could sort through hundreds of CVs within minutes, ensuring unbiased and consistent evaluation based on predefined criteria. The outcome of this project proves the potential of combining Agentic AI with LLMs in solving real-world HR challenges. It lays the foundation for future applications in automated candidate shortlisting, personalized interview question generation, and overall recruitment intelligence. The project showcases how AI can empower human decision-making rather than replace it, making hiring more efficient, scalable, and fair.