Skip to content

Home

Jump to bottom
Dwarkesh edited this page Dec 25, 2025 · 1 revision

AI-Hackathon-Judge Wiki ⚖️🤖

Welcome to the official Wiki for AI-Hackathon-Judge — an AI-powered platform designed to evaluate hackathon projects automatically using repository analysis and demo video evaluation.

📌 Table of Contents

  1. Introduction
  2. Problem Statement
  3. Solution Overview
  4. Features
  5. System Architecture
  6. Tech Stack
  7. Installation Guide
  8. Usage Guide
  9. Evaluation Criteria
  10. Security Considerations
  11. Limitations
  12. Future Enhancements
  13. Contribution Guidelines
  14. License

1 : Introduction

AI-Hackathon-Judge is an intelligent judging system that analyzes GitHub repositories and project demo videos to provide instant scores and structured feedback for hackathon submissions.

It minimizes human bias, saves judging time, and ensures consistent evaluation across all participants.

2 : Problem Statement

Traditional hackathon judging faces multiple challenges:

  • Manual evaluation is time-consuming
  • Human bias affects fairness
  • Judges may not fully review large codebases
  • Feedback is delayed or insufficient
  • Technical depth is often overlooked

3 : Solution Overview

AI-Hackathon-Judge automates the judging process by:

  • Fetching and analyzing GitHub repositories
  • Evaluating demo videos (link-based)
  • Using AI models to assess code quality, innovation, and documentation
  • Generating structured scores and feedback
  • Providing instant results via a web interface

4 : Features ✨

  • 🔍 GitHub Repository Analysis
  • 🎥 Demo Video Evaluation
  • 🤖 AI-Based Scoring Engine
  • 📊 Instant Results & Feedback
  • 🧠 Bias-Free Judging
  • ⚡ Fast & Scalable Architecture
  • 🌐 Web-Based Interface
  • 💻 Optional Desktop Executable Support

5 : System Architecture 🏗️

Frontend (React + Vite)
        ↓
FastAPI Backend (Python)
        ↓
AI Model (Gemini API)
        ↓
Scoring & Feedback Engine

Components

  • Frontend – User interface for input and result visualization
  • Backend API – Handles logic, processing, and AI interaction
  • AI Engine – Performs analysis and scoring
  • GitHub API – Fetches repository data

6 : Tech Stack 🧑‍💻

Frontend

  • React
  • Vite
  • Tailwind CSS
  • Axios

Backend

  • Python
  • FastAPI
  • Uvicorn

AI & Tools

  • Google Gemini API
  • GitHub REST API
  • PyInstaller (for local builds)

7 : Installation Guide ⚙️

Backend Setup

git clone https://github.com/Daku3011/AI-Hackathon-Judge
cd backend
pip install -r requirements.txt

Create a .env file:

GEMINI_API_KEY=your_api_key_here

Run the backend server:

uvicorn main:app --reload

Frontend Setup

cd frontend
npm install
npm run dev

8 : Usage Guide 🚀

  1. Open the web interface
  2. Enter:
    • GitHub repository URL
    • Demo video link (optional)
  3. Click Evaluate
  4. AI processes the submission
  5. View:
    • Final score
    • Category-wise analysis
    • Strengths & weaknesses
    • AI-generated feedback

9 : Evaluation Criteria 📊

Category Description
Code Quality Readability, structure, best practices
Innovation Creativity and uniqueness
Documentation README clarity and setup instructions
Feasibility Practicality and real-world usability
Presentation Demo explanation and clarity

10 : Security Considerations

  • API keys are stored using environment variables
  • No sensitive user data is stored
  • Repositories are accessed in read-only mode
  • Public deployment should include rate limiting
  • Refer to SECURITY.md for vulnerability reporting

11 : Limitations ⚠️

  • AI judgment depends on prompt and model behavior
  • Video evaluation is limited to metadata and transcripts
  • Private repositories require user-provided access
  • Incomplete documentation may affect scoring accuracy

12 : Future Enhancements 🔮

  • Admin & judge dashboards
  • Multi-AI consensus scoring
  • Plagiarism detection
  • Team-based evaluation
  • PDF scorecard export
  • Authentication & role management
  • Cloud deployment templates

13 : Contribution Guidelines 🤝

Contributions are welcome!

  1. Fork the repository
  2. Create a feature branch
  3. Write clean, documented code
  4. Commit with clear messages
  5. Submit a pull request

14 : License 📄

This project is licensed under the MIT License.
You are free to use, modify, and distribute it with proper attribution.

⭐ Final Note

AI-Hackathon-Judge aims to modernize hackathon judging by making it faster, smarter, and fairer using AI.

Clone this wiki locally