International Journal of Scientific Research and Engineering Development

In today’s job market, students and fresh graduates face a huge challenge when it comes to preparing for interviews. Most of them practice on their own or rely on friends for mock interviews, which does not give them any real feedback on how they actually come across during an interview. There is a clear gap between what candidates know and how well they perform under pressure. PrepMate is a web-based platform developed to solve this problem by combining subject-knowledge quizzes, AI-generated mock interviews, real-time emotion detection, and detailed analytics — all in one place. The system is built using React and TypeScript on the frontend, with Supabase handling the backend database and authentication. For AI, the Google Gemini API is used to generate quiz questions and interview questions tailored to a user’s profile. The emotion detection is implemented using the face-api.js library, which runs directly in the browser, meaning no data is sent to any external server for face analysis. The platform has four main modules: a Mock Quiz module, a Mock Interview module, an Analytics module, and the Emotion Detection feature integrated into the interview.

Tanvi S. Jivatode, Nishant M. Kadu, Astha M. Bhande, Anandi A. Thakare, Prof. Shraddha N. Utane,"PrepMate: AI Based Interview Preparation Platform" International Journal of Scientific Research and Engineering Development, V9(2): Page(2233-2237) Mar-Apr 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.