PrepPilot – AI Mock Interview Platform

A full-stack AI-powered mock interview platform that generates role-specific interview questions using Gemini AI, evaluates your answers, and provides structured feedback with scoring.

Problem It Solves

Most interview preparation is passive — reading answers, watching videos. PrepPilot makes it active. You describe the role, get tailored questions, answer them (by voice or text), and receive AI feedback on each answer instantly.

Tech Stack

Backend

• Java 21 + Spring Boot 3.5
• Spring Security + JWT Authentication
• Spring Data JPA + Hibernate
• PostgreSQL
• Gemini AI API (google-generativelanguage)

Frontend

• React 18 + Vite
• Tailwind CSS
• Axios
• React Router DOM

Features

• User registration and login with JWT-based authentication
• Create mock interviews by specifying job position, description, and experience level
• AI-generated interview questions tailored to the role
• Answer via voice (Web Speech API) or text
• Per-answer AI feedback with rating out of 10
• Dashboard showing all past interviews
• Feedback page with average score across all answers

Project Structure

PrepPilot/

├── backend/# Spring Boot REST API
│   └── src/main/java/com/preppilot/backend/
│       ├── controller/
│       ├── service/
│       ├── model/
│       ├── repository/
│       ├── security/
│       ├── dto/
│       └── exception/
└── frontend/# React + Vite
└── src/
├── pages/
└── api/

Local Setup

Prerequisites

• Java 21
• Node.js 18+
• PostgreSQL
• Gemini API key (aistudio.google.com)

Backend

1. Clone the repo
2. Create PostgreSQL database:

CREATE DATABASE preppilot;

3. Run the schema:

CREATE TABLE users (
    id BIGSERIAL PRIMARY KEY,
    name VARCHAR(100) NOT NULL,
    email VARCHAR(150) UNIQUE NOT NULL,
    password VARCHAR(255) NOT NULL,
    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);

CREATE TABLE mock_interviews (
    id BIGSERIAL PRIMARY KEY,
    user_id BIGINT REFERENCES users(id) ON DELETE CASCADE,
    job_position VARCHAR(100) NOT NULL,
    job_description TEXT NOT NULL,
    experience_level VARCHAR(50) NOT NULL,
    questions_json TEXT NOT NULL,
    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);

CREATE TABLE user_answers (
    id BIGSERIAL PRIMARY KEY,
    interview_id BIGINT REFERENCES mock_interviews(id) ON DELETE CASCADE,
    question TEXT NOT NULL,
    user_answer TEXT,
    ai_feedback TEXT,
    ai_rating INTEGER,
    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);

4. Copy backend/src/main/resources/application.properties.example to application.properties and fill in your values
5. Run:

cd backend
./mvnw spring-boot:run

Backend starts on http://localhost:8080

Frontend

cd frontend
npm install
npm run dev

Frontend starts on http://localhost:5173

API Endpoints

Method	Endpoint	Auth	Description
POST	/api/auth/register	No	Register new user
POST	/api/auth/login	No	Login and get JWT
POST	/api/interviews	Yes	Create interview + generate questions
GET	/api/interviews	Yes	Get all user interviews
GET	/api/interviews/{id}	Yes	Get single interview
POST	/api/interviews/answer	Yes	Submit answer + get AI feedback
GET	/api/interviews/{id}/answers	Yes	Get all answers for interview

Environment Variables

See backend/src/main/resources/application.properties.example for all required configuration values.

Author

Bhawani Shanker Sharma LinkedIn | GitHub