International Journal of Scientific Research and Engineering Development

This paper presents an intelligent web-based interview preparation platform that integrates the MERN stack (MongoDB, Express, React, Node.js) with the Gemini API to deliver adaptive mock interviews and AI-driven evaluation. The system aims to simulate realistic job inter- view environments, provide dynamic feedback, and assess candidates across communication, knowledge, and fluency metrics. The platform ensures scalability, low latency, and cost-effective API usage through caching and heuristic pre-processing. A pilot evaluation with four undergraduate students demonstrated measurable improvements in performance, confidence, and system responsiveness. The paper discusses architecture, implementation, results, and future enhancements for extending the system to speech and multimodal analysis.