AI INTERVIEWER

Inspiration

Many students and job seekers struggle with technical interviews because they lack a realistic, high-pressure environment to practice in. We wanted to bridge this gap by creating an accessible, intelligent mock interviewer that feels just like sitting across from a real hiring manager.

What it does

Ram Sir conducts a fully voice-driven, real-time interview. Candidates input their name, target role, and experience level, and the AI generates tailored questions on the spot. Using speech recognition, the platform listens to the candidate's answers and evaluates them in real-time based on technical accuracy, communication skills, and confidence. Afterward, it generates a comprehensive analytics dashboard and a downloadable PDF report highlighting strengths and areas for improvement.

How we built it

Frontend: We built a highly responsive, type-safe interface using React 19, TypeScript, and Vite. UI/UX: The sleek "glassmorphism" aesthetic was crafted using Tailwind CSS v4, Lucide React for icons, and Framer Motion for fluid animations. Data Visualization: The post-interview dashboard uses Recharts to render performance radar and bar charts. AI Engine: The core intelligence is powered by the Google Gemini SDK (@google/genai). We utilize the gemini-3.1-flash-lite model for high-throughput, low-latency prompt generation and evaluation. Voice Integration: We leveraged native browser SpeechRecognition and SpeechSynthesis APIs to handle the spoken back-and-forth.

Challenges we ran into

The biggest technical hurdle was seamlessly connecting the browser's native speech recognition with the Gemini AI. We had to ensure the app accurately captured live audio, converted it to text, sent it to the AI, and generated the next spoken question without awkward delays. We also had to navigate API rate limits and optimize our request latency by migrating to a lightweight "Flash" model.

Accomplishments that we're proud of

We successfully built a hands-free, end-to-end interview loop that provides immediate, actionable feedback. We are especially proud of the comprehensive analytics dashboard, which breaks down a candidate's performance into clear, quantifiable metrics that genuinely help them improve.

Built With

Submitted to

Department Battle: Hack Days

Created by

I designed, developed, and deployed the AI Interviewer platform from the ground up. I architected the frontend systems, creating a seamless end-to-end interview experience for candidates and recruiters. The platform leverages advanced AI technologies to conduct automated interviews, dynamically generate contextual questions, analyze candidate responses, and provide detailed evaluation reports. I implemented the full interview workflow, including user authentication, interview creation and management, candidate onboarding, real-time conversational interactions, transcript generation, response processing, and recruiter dashboards. The application was built using modern web technologies including React, Next.js, TypeScript, Firebase, and Google Gemini AI, ensuring scalability, responsiveness, and reliability across devices. I engineered the AI orchestration layer responsible for question generation, conversational flow management, response evaluation, and feedback generation, enabling interviews to adapt naturally based on candidate answers. Additionally, I integrated speech-processing capabilities, cloud infrastructure, database management, and secure data handling mechanisms to create a production-ready platform. Beyond the core functionality, I focused on user experience, performance optimization, system architecture, deployment pipelines, and overall product design. This project required combining expertise across full-stack development, artificial intelligence, prompt engineering, cloud services, database design, authentication systems, and human-centered product development to deliver a comprehensive AI-powered recruitment solution that improves hiring efficiency, consistency, and scalability. The platform demonstrates my ability to transform a complex idea into a fully functional product by managing every stage of the software development lifecycle, from concept and architecture design to implementation, testing, deployment, and continuous improvement. The resulting system provides organizations with an intelligent interviewing solution capable of conducting structured assessments, generating standardized evaluations, and streamlining recruitment workflows through automation and AI-driven decision support.

Samarth Khatri
adityabijore6024-del bijore

Updates

adityabijore6024-del bijore started this project — Jun 18, 2026 07:10 AM EDT

Leave feedback in the comments!

Log in or sign up for Devpost to join the conversation.