Prepverse.AI

Home page
Interview section with face detection
Resume building page
Quiz page

Inspiration

We all know how stressful interviews can be — not just for candidates, but also for recruiters. As students, we often struggled with preparing for real interviews, and we realized companies also face difficulties in fairly evaluating resumes and candidates. Many students also don’t get enough opportunities to practice interviews in a realistic way, which adds to their anxiety. This inspired us to build Prepverse.AI, a platform that not only makes interviews more interactive and resumes more effective, but also provides a dedicated mock interview section for students to practice and build confidence before facing real recruiters.

What it does

Prepverse.AI is an end-to-end solution for interview preparation and recruitment.

Candidates can attend AI-powered interviews that ask dynamic, personalized questions based on their resumes.
The platform uses voice interaction and face detection to simulate a real interview experience.
Resumes are parsed and scored with an ATS-like system, giving candidates clear insights into their strengths and gaps.
It also provides an AI-assisted resume builder, helping job seekers improve their chances.
On the other side, companies get a secure dashboard to create AI-driven interviews, review candidate performance, and simplify the hiring process.

How we built it

We combined several technologies to bring Prepverse.AI to life:

React.js + Tailwind CSS for the frontend, ensuring a clean and responsive interface.
Flask and Express.js as backend services to handle AI tasks and API communication.
MongoDB for storing resumes, candidate data, and company interview setups.
Google Gemini API to generate interview questions, analyze answers, and provide instant feedback.
MediaPipe to track face movements and eye contact during interviews, making the simulation more realistic.

Challenges we ran into

Building Prepverse.AI wasn’t easy. Some of the toughest challenges were:

Making different frameworks (React, Flask, Express) work smoothly together.
Ensuring real-time voice-to-text accuracy for candidate answers.
Designing a reliable resume parsing + ATS scoring system that feels meaningful to candidates.
Managing secure role-based access for candidates and recruiters.

Accomplishments that we're proud of

Creating a smart AI interviewer that adapts to what the candidate says.
Integrating both resume evaluation and resume building in one place.
Delivering a company portal that gives recruiters real insights, not just raw data.
Successfully using MediaPipe for engagement tracking, something that really adds to the experience.

What we learned

This project taught us a lot:

How to combine multiple AI tools (LLMs + vision models) into a single workflow.
The importance of scalability and clean system design, especially for real-time apps.
New technical skills in MERN stack development, Flask integration, and API orchestration.
A much better understanding of how ATS scoring and resume evaluation actually work in industry.

What's next for Prepverse.AI

We’re excited about where this can go. Some next steps include:

Adding multilingual support so candidates worldwide can use it.
Building an AI recommendation engine to guide candidates on how to improve their skills.