This paper presents InterviewAI, a Software-as-a-Service (SaaS) platform that conducts live mock interviews using artificial intelligence. Candidates often lack access to affordable and realistic interview practice, and existing systems either rely on static question banks or cover only parts of the interview experience. In InterviewAI, a user signs in with Google Authentication and uploads a resume in PDF format. The system extracts skills, education and experience, generates personalised questions, conducts the interview through real-time voice and video conversation with an AI interviewer, and produces a performance report covering strengths, weaknesses, communication skills and suggestions for improvement. A credit-based usage model with Razorpay payment integration makes the platform suitable for production use. The system is built on React.js, Node.js, Express.js and MongoDB, with Firebase authentication, Razorpay payments and OpenRouter for AI model access, and is deployed on Render. The platform was tried end to end with a few users to check its working, and all core modules, from sign-in to feedback report and payment, worked as intended.

Ananya Rapolu, Manoj Singh, "InterviewAI: An Intelligent Virtual Interview Platform with Real-Time Voice, Video, and AI-Based Performance Evaluation" International Journal of Scientific Research and Engineering Development, V9(5): Page(848-853) September - October 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.