Code Clasher

AWS-based programming puzzles platform with LLM-enhanced test suite generation system, Next.js + Vercel frontend, and a secure & scalable pipeline for execution of arbitrary user code.

Demo (Archived): https://sse-team-project.vercel.app
GitHub Repository: https://github.com/AndyXIP/sse-team-project

---

Overview

Code Clasher is a competitive programming platform where developers solve daily coding challenges, compete on global leaderboards, and sharpen their algorithmic skills. Built as a Software Systems Engineering group project at Imperial College London, this platform showcases modern cloud architecture, real-time features, and secure code execution.

Key Features

• Daily Challenges: Fresh easy/hard coding problems with LLM-generated test cases
• Global Leaderboard: Real-time rankings and progress tracking
• In-Browser Code Editor: Syntax highlighting, multiple language support
• Real-Time Execution: Secure sandboxed code evaluation with instant feedback
• Progress Analytics: Personal stats, streaks, and performance insights
• Modern UI: Dark/light mode, responsive design, seamless UX

---

Architecture

┌─────────────────┐    ┌──────────────────┐    ┌─────────────────┐
│   Next.js       │    │   FastAPI        │    │   AWS Lambda    │
│   Frontend      │◄──►│   Main API       │◄──►│   Microservices │
│   (Vercel)      │    │   (Elastic       │    │   (Code Eval,   │
└─────────────────┘    │   Beanstalk)     │    │   Leaderboard)  │
                       └──────────────────┘    └─────────────────┘
                                │                       │
                       ┌────────▼──────────┐    ┌──────▼──────────┐
                       │   Valkey Redis    │    │   Supabase      │
                       │   (Caching)       │    │   (PostgreSQL)  │
                       └───────────────────┘    └─────────────────┘

Technology Stack

Frontend:

• Next.js 14 with App Router
• TypeScript & Tailwind CSS
• React with modern hooks
• Vercel deployment

Backend Services:

• FastAPI (Main API server)
• AWS Lambda (Serverless microservices)
• Python 3.9+ with async/await
• AWS SQS (Message queuing)
• AWS Elastic Beanstalk (Container deployment)

Data & Caching:

• Supabase (PostgreSQL database)
• Valkey Glide (Redis-compatible caching)
• HuggingFace Datasets (Problem source)

Infrastructure:

• Docker containerization
• AWS cloud services
• GitHub Actions (CI/CD ready)

---

Quick Start

Prerequisites

• Node.js 18+
• Python 3.9+
• Docker (optional)
• AWS CLI (for deployment)

Local Development

1. Clone the repository

   git clone https://github.com/AndyXIP/sse-team-project.git
   cd code-clasher

2. Frontend Setup

   cd frontend
   npm install
   npm run dev

3. Backend Setup

   cd main-api
   pip install -r requirements.txt
   uvicorn app:app --reload

4. Environment Variables

   # Create .env files in respective directories
   cp .env.example .env
   # Configure AWS credentials, database URLs, etc.

---

Project Structure

code-clasher/
├── frontend/                 # Next.js frontend application
├── main-api/                 # FastAPI central API server
├── lambda-code-evaluator-v2/ # Serverless code execution engine
├── leaderboard/              # Leaderboard processing service
├── questions/                # Question fetching & caching
├── leaderboard-updater/      # Real-time leaderboard updates
├── test-case-generation/     # LLM-powered test case creation
└── README.md                 # This file

Each service includes its own comprehensive README with detailed setup instructions.

---

How It Works

1. Daily Questions: LLM generates test cases for coding problems from HuggingFace dataset
2. Code Submission: Users write solutions in the browser-based editor
3. Secure Execution: Code runs in isolated AWS Lambda containers
4. Real-Time Results: WebSocket updates provide instant feedback
5. Leaderboard Updates: Successful submissions update global rankings
6. Progress Tracking: User stats and streaks are maintained across sessions

---

Testing & Quality

• Unit Tests: Comprehensive test suites for all services
• Integration Tests: End-to-end API testing with mocked AWS services
• Security: Input validation, sandboxed execution, rate limiting
• Performance: Caching strategies, async processing, optimized queries

---

Deployment

Production URLs (Archived):

• Frontend: https://sse-team-project.vercel.app (No longer active)
• Main API: https://main-api.click (No longer active)

Infrastructure:

• Vercel (Frontend hosting)
• AWS Elastic Beanstalk (API server)
• AWS Lambda (Microservices)
• Supabase (Managed PostgreSQL)

---

Team

Developed as a Software Systems Engineering group project at Imperial College London by:

• Andy
• Lucas
• Will
• Yile

---

License

This project is licensed under the MIT License - see the LICENSE file for details.

---

🔗 Links

• Live Application (Archived): https://sse-team-project.vercel.app
• GitHub Repository: https://github.com/AndyXIP/sse-team-project
• Problem Dataset: HuggingFace APPS Dataset

---

Built at Imperial College London