AI Interview Bot for Freshers Using Google Gemini and LangChain

Abstract

This paper introduces an AI-driven web application 
for automated interview evaluation, built using 
Django and integrated with Google Gemini through 
LangChain. It streamlines the recruitment process by 
generating role-specific interview questions and 
assessing candidate responses in real time, making it 
especially suitable for fresh graduates and entry-level 
roles. Candidates register via a secure portal and 
undergo a four-question interview tailored to their 
selected 
job role. Questions are generated 
dynamically using a large language model (LLM), 
ensuring relevance and simplicity. Each response is 
evaluated by the AI, which returns a score (0–5) and 
a qualification status in structured JSON format. At 
the end of the session, the system calculates the 
average score, determines the final result, and 
automatically emails it to the candidate. The platform 
includes admin features for managing users and 
viewing results, along with OTP-based password 
recovery and profile image support. This scalable, 
unbiased system offers an efficient solution for bulk 
hiring and academic assessments by minimizing 
human effort while maintaining evaluation accuracy.