AI Technical Interview Generator

A full-stack web application powered by the MERN stack and integrated with Gemini AI, designed to help users simulate and prepare for technical interviews. This tool generates a personalized list of potential interview questions and concise answers based on the position's experience requirements and the selected discussion topics.

🔒 Authentication is required to access the main features. Users can sign up or log in to start generating sessions.

---

🌟 Features

• 🔐 Authentication — Register & login to access core features.
• 🧠 AI-powered Question Generation — Generate 10 technical interview questions and concise answers based on:
  • Position
  • Years of experience
  • Topics of discussion
  • Notes
• 📌 Pin Important Questions — Mark important questions and move them to the top.
• 📝 Add More Questions — Add more questions to the session.
• 📖 Detailed Explanation — View deeper explanations to better understand the context of each question.
• 🗂️ Session Management — Create, delete, and explore session details.
• 🐳 Docker support for quick deployment

---

🛠️ Built With

Layer	Stack
Frontend	React (Vite) + Tailwind CSS 4
Backend	Express.js + Node.js v22.15.0
Database	MongoDB
Package Manager	npm
AI Model	Gemini (via API integration)
Container	Docker + Docker Compose

---

⚙️ Setup & Installation

You can run the project in two ways:

🧩 Option 1: Using Docker (Recommended)

Make sure you have Docker installed.

🧪 Requirements

• Docker
• Gemini API Key (get it from Gemini Studio)
• JWT Secret (generate using the command below)

node -e "console.log(require('crypto').randomBytes(64).toString('hex'))"

▶️ Steps

1. Clone this repository:

git clone https://github.com/dewangga-pk/sd-ai-tech-interview.git
cd sd-ai-tech-interview

2. Open docker-compose.yml and replace the following placeholders:

JWT_SECRET: <your_jwt_secret>
GEMINI_API_KEY: <your_gemini_api_key>

3. Run the app:

docker-compose up --build

4. Open in your browser: http://localhost:3000

---

🛠️ Option 2: Manual Setup (No Docker)

Requirements:

• Node.js v22.15.0 → Download Node.js
• MongoDB → Install MongoDB

Steps:

1. Clone the repository:

git clone https://github.com/dewangga-pk/sd-ai-tech-interview.git

2. Setup Backend:

• Go to backend/
• Copy .env.example to .env
• Fill the .env file with:

MONGO_URI=mongodb://localhost:27017/interview_tool
JWT_SECRET=<your_jwt_secret>
GEMINI_API_KEY=<your_gemini_api_key>
PORT=8000

Generate a JWT secret with:

node -e "console.log(require('crypto').randomBytes(64).toString('hex'))"

3. Setup Frontend:

• Go to frontend/
• Copy .env.example to .env
• Set VITE_API_BASE_URL to your backend URL, e.g.

VITE_API_BASE_URL=http://localhost:8000

4. Install dependencies and run both frontend & backend:

Terminal 1: Backend

cd backend
npm install
npm run dev

Terminal 2: Frontend

cd frontend
npm install
npm run dev

By default, the frontend will be served at: http://localhost:5173

---

📸 Screenshots

---

📁 Folder Structure

sd-ai-tech-interview/
├── backend/           # Express backend (API, DB connection, auth, routes)
├── frontend/          # React (Vite) frontend (UI, session forms, question display)
├── docker-compose.yml
├── README.md

---

🛡️ Security Notice

• This project uses .env files for all sensitive configs
• Actual .env files are ignored via .gitignore
• Please use the .env.example as a reference
• Do not commit API keys, secrets, or database credentials