Skip to content

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

11 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

πŸš€ JudgeLens

AI-Powered Hackathon Evaluation Platform

JudgeLens is an AI-powered judging assistant designed to help hackathon organizers and judges evaluate submissions faster, more consistently, and at scale.

Instead of manually reviewing dozens or hundreds of repositories, pitch decks, and UI designs, JudgeLens automatically analyzes submissions and generates structured evaluations, rankings, and actionable feedback.


🎯 The Problem

Modern hackathons often receive hundreds of submissions.

Judges typically need to review:

  • GitHub repositories
  • Pitch decks
  • UI/UX designs
  • Business models
  • Technical implementations

This process is:

  • ⏳ Time-consuming
  • πŸ“ˆ Difficult to scale
  • βš–οΈ Subjective and inconsistent
  • 😡 Prone to evaluator fatigue

JudgeLens helps solve this challenge using AI-powered evaluation pipelines.


✨ Features

πŸ“‚ Repository Intelligence Engine

Analyze GitHub repositories for:

  • Documentation Quality
  • Technical Complexity
  • Project Completeness
  • Repository Structure
  • Development Practices

Outputs:

  • Repository Score
  • Strengths
  • Weaknesses
  • Recommendations

πŸ“Š Pitch Deck Evaluation

Automatically reviews uploaded presentations and scores:

  • Innovation
  • Technical Complexity
  • Business Impact
  • Presentation Quality

Generates detailed feedback and recommendations.

🎨 UI/UX Evaluation

Analyze application screenshots using AI vision.

Measures:

  • Visual Design
  • UX Quality
  • Professionalism
  • Interface Clarity

Provides actionable UI improvement suggestions.

πŸ† Dynamic Leaderboard

Projects are automatically ranked based on:

  • Repository Evaluation
  • Presentation Evaluation
  • UI/UX Evaluation

Allowing judges to identify top-performing projects instantly.

🧠 AI-Powered Verdicts

JudgeLens doesn't just score projects.

It explains:

  • Why a project received its score
  • Key strengths
  • Critical weaknesses
  • Strategic recommendations

πŸ“Έ Screenshots

πŸ† Leaderboard Dashboard

image

πŸ“€ Project Submission Portal

image

πŸ“ˆ Evaluation Dashboard

image

πŸ” Detailed Project Analysis

image image image

βš™οΈ How It Works

Project Submission
        ↓
Repository Analysis
        ↓
Pitch Deck Analysis
        ↓
UI Screenshot Analysis
        ↓
AI Evaluation Engine
        ↓
Scoring & Ranking
        ↓
Leaderboard & Reports

πŸ—οΈ System Architecture

Frontend (React + Tailwind)
            ↓
       FastAPI Backend
            ↓
      Google Gemini
            ↓
      MongoDB Atlas
            ↓
 Project Evaluations

πŸ› οΈ Tech Stack

Frontend

  • React
  • Vite
  • Tailwind CSS
  • shadcn/ui
  • Recharts

Backend

  • FastAPI
  • Python

Database

  • MongoDB Atlas

Artificial Intelligence

  • Google Gemini 2.5 Flash
  • Vertex AI

Cloud

  • Google Cloud Platform (GCP)

Integrations

  • GitHub API
  • PyMongo
  • python-pptx

πŸš€ Local Setup

Clone Repository

git clone https://github.com/JVikramV/JudgeLens.git

cd JudgeLens

Backend Setup

cd backend

python -m venv venv

source venv/bin/activate

pip install -r requirements.txt

uvicorn main:app --reload

Backend runs at:

http://localhost:8000

Frontend Setup

cd frontend

npm install

npm run dev

Frontend runs at:

http://localhost:5173

πŸ”‘ Environment Variables

Create:

backend/.env
MONGO_URI=your_mongodb_connection_string

DATABASE_NAME=judgelens

PROJECT_ID=your_google_cloud_project

LOCATION=us-central1

GITHUB_TOKEN=your_github_personal_access_token

πŸ“Š Evaluation Categories

Category Metrics
Repository Documentation, Technical Quality, Completeness
Pitch Deck Innovation, Business Impact, Technical Complexity
UI/UX Visual Design, User Experience, Professionalism

🎯 Why JudgeLens?

Traditional Hackathon Judging:

  • Manual repository review
  • Manual PPT evaluation
  • Manual UI assessment
  • Subjective scoring
  • Difficult to scale

JudgeLens:

βœ… Automated Evaluation

βœ… Consistent Scoring

βœ… Detailed Feedback

βœ… Dynamic Ranking

βœ… AI-Powered Insights

βœ… Scalable Judging


πŸ“ˆ Future Roadmap

Phase 2

  • Demo Video Analysis
  • Deep Repository Inspection
  • AI Judge Agents
  • Team Evaluation

Phase 3

  • Multi-Agent Judging
  • Fraud Detection
  • Automatic Finalist Selection
  • Investor Readiness Scoring
  • AI-Powered Winner Prediction

πŸ‘¨β€πŸ’» Author

V Jayanth

Built for innovation, learning, and the future of AI-assisted evaluation.


πŸ“œ License

MIT License


⭐ If you found JudgeLens interesting, consider starring the repository.

About

JudgeLens autonomously reviews submissions,checks compliance,generates evaluation reports,stores results,maintains rankings,and assists judges in selecting finalists.

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages