Mainframe

Advanced AI Interview Simulator powered by Google Gemini.

Overview

Mainframe is a high-precision career readiness platform designed to simulate professional interviews with real-time evaluation. Utilizing A.R.I.A (Adaptive Response Interface Agent), the system provides an automated, data-driven environment for mastering interview performance across various sectors and roles.

Core Capabilities

• Intelligent Question Generation: Dynamic MCQ engine capable of simulating Technical, HR, and Situational interviews for over 20 professional roles.
• On-Demand Deep Analysis: Real-time evaluation of user responses using sophisticated AI models to provide precise educational feedback.
• Performance Intelligence: Detailed tracking of accuracy metrics, historical trends, and category-specific performance data.
• Adaptive Learning Systems: Personalized study roadmaps generated based on individual performance history to target specific areas for improvement.
• Persistent Architecture: Secure session management ensuring progress is maintained across local environments.

Technical Architecture

• Interface: React 19, TypeScript, Vite, Tailwind CSS.
• Core Engine: Flask (Python), SQLite3.
• Intelligence: Google Gemini 3.1 Flash Lite API.
• Design Philosophy: Minimalist, editorial-style aesthetic focusing on clarity and professional focus.

Installation and Execution

One-Click Automation (Recommended)

To launch the backend, frontend, and an ngrok tunnel (for internet access) simultaneously, use the provided startup script:

chmod +x run.sh
./run.sh

This script will automatically clear ports, launch all services, and provide you with a live public link via ngrok.

Manual Configuration

Backend Configuration

1. Initialize the Python environment:

   python3 -m venv venv
   source venv/bin/activate
   pip install -r requirements.txt

2. Configure environment variables in a .env file:

   GEMINI_API_KEY=your_api_key

3. Initiate the API server:

   python3 app.py

Frontend Configuration

1. Install dependencies and start the development server:

   npm install
   npm run dev

The platform will be accessible at http://localhost:5173.

Licensing

Distributed under the MIT License. See LICENSE for more information.

Acknowledgments

• Motion Sites: Special recognition to Motion Sites for providing essential animation resources and aesthetic direction.
• Google Generative AI: Powered by the Gemini family of models for high-speed, accurate simulations.

---

Mainframe is developed and maintained by Rahul-encc.