AI INTERVIEWER

AI Interviewer is web app that conducts real-time video interviews using AI, analyzes answers, behavior, prevents cheating, and delivers detailed interview reports for candidates and companies.

Comment

Inspiration

Interviews are one of the most important yet unfair stages of recruitment and selection. Many candidates face biased evaluations, inconsistent interview standards, limited access to quality interview practice, and stress caused by human judgment. This problem becomes even more serious for high-stakes interviews such as technical roles, campus placements, and competitive examinations like UPSC.

We were inspired to create AI INTERVIEWER to make interviews more accessible, unbiased, and scalable using AI. Our goal was to simulate a real interview environment while maintaining fairness, integrity, and professionalism.

What it does

AI INTERVIEWER is a web-based platform that conducts real-time AI-powered video interviews.

  • Candidates select the role they are preparing for (technical, mechanical, management, UPSC, etc.)
  • The AI dynamically asks 10–15 interview questions using audio and video, not text
  • The system monitors facial focus, audio consistency, and screen sharing to prevent cheating
  • Companies can host interviews by creating time slots and sharing a secure interview code
  • After completion, the AI generates an evaluation report and securely stores interview recordings
  • A Practice Mode allows candidates to prepare before taking a real interview

How we built it

We built AI INTERVIEWER using a modern, scalable tech stack:

  • Frontend: React / Next.js for a clean and professional UI
  • Backend: Node.js / FastAPI for handling sessions and logic
  • AI: Google Gemini 3 for question generation, response analysis, and evaluation
  • Video & Audio: WebRTC for real-time video interviews
  • Speech Processing: Google Speech-to-Text and Text-to-Speech
  • Computer Vision: Face and focus detection using Google Vision APIs
  • Authentication & Cloud: Google Cloud and Firebase for secure access and storage

The AI adapts questions based on user profile, interview type, and response quality.

Challenges we ran into

  • Implementing real-time video interviews with stable audio and low latency
  • Ensuring anti-cheating measures without violating user privacy
  • Designing an AI interviewer that feels professional and non-intimidating
  • Managing permissions for camera, microphone, and screen sharing smoothly
  • Balancing fairness, ethics, and automation in evaluation

Accomplishments that we're proud of

  • Built a fully AI-driven video interview experience
  • Successfully integrated Gemini 3 for adaptive questioning
  • Created a dual-mode system (Practice + Real Interview)
  • Designed a secure company-hosted interview workflow
  • Implemented behavioral monitoring while maintaining transparency

What we learned

Through this project, we learned how to:

  • Design AI systems that interact with humans in real time
  • Integrate multiple Google AI services into one platform
  • Handle ethical considerations in AI-based assessment
  • Build scalable, interview-grade applications under time constraints
  • Work effectively as a team on a complex AI product

What's next for AI INTERVIEWER

In the future, we plan to:

  • Add multilingual interview support
  • Improve emotion and sentiment analysis
  • Introduce resume-based personalized interviews
  • Enable enterprise-level dashboards for companies
  • Expand use cases to education, certification, and government exams

Built With

Share this project:

Updates