CodePair is a production-grade, real-time collaborative coding interview platform designed to simulate realistic technical interviews. It seamlessly matches peers based on skill level and provides a synchronized IDE environment for solving algorithmic problems together.
🔴 Live Demo: Try CodePair Now
Ever wanted to remove that interview anxiety?
Technical interviews are stressful. The only way to get comfortable is to practice exactly how you play. CodePair isn't just a shared editor; it's a simulation engine for your next big interview.
- Conquer the Nerves: Get used to explaining your thought process in real-time while coding in front of a peer.
- Find Your Match: Stop practicing alone. Our smart queue pairs you with partners at your exact skill level.
- Master the Format: From "Easy" warm-ups to "Hard" dynamic programming, practice the full range of algorithmic challenges.
- Intelligent Peer Matching: Instant, skill-based matchmaking queue that pairs users by preferred language (Go, Python, Java, JS/TS) and problem difficulty.
- Real-Time Collaboration: Low-latency code synchronization allowing two users to type, edit, and debug simultaneously in the same environment.
- Production-Grade Editor: Integrated Monaco Editor (the core of VS Code) for a familiar, powerful coding experience with syntax highlighting and auto-completion.
- LeetCode Integration: Dynamically fetches algorithmic problems from LeetCode or falls back to a curated local repository.
- Interactive Feedback System: robust session tracking, post-interview feedback, and history management.
CodePair is built with a modern, high-performance stack ensuring scalability and responsiveness.
- Language: Go (Golang) 1.22 - Chosen for high concurrency and low latency.
- Framework: Fiber - Express-inspired web framework, optimized for speed.
- Real-Time: WebSocket - Full-duplex communication for state synchronization and match notifications.
- Database: PostgreSQL 16 - Relational data integrity for users, sessions, and problems.
- Caching & Pub/Sub: Redis 7 - Ephemeral state management and distributed message passing.
- Framework: React 18 + TypeScript - Type-safe, component-based UI.
- Build Tool: Vite - Next-generation frontend tooling.
- Styling: CSS Modules + Vanilla CSS Variables - Scoped, performant styling without style leakage.
- State Management: Zustand - Minimalist state management.
- Editor: Monaco Editor - Industry-standard code editor implementation.
- Containerization: Docker - Multi-stage builds for optimized production images.
- IaC: Terraform - Declarative infrastructure management.
- Cloud Provider: AWS
- Compute: ECS Fargate (Serverless Containers)
- Load Balancing: Application Load Balancer (ALB)
- Database: RDS for PostgreSQL
- Cache: ElastiCache for Redis
- Registry: Amazon ECR
- CI/CD: GitHub Actions - Automated pipelines for linting, testing, and deployment checks.
- Go 1.22+
- Node.js 20+
- Docker & Docker Compose
- AWS CLI & Terraform (for deployment)
-
Clone and Setup
git clone https://github.com/yourusername/codepair.git cd codepair cp .env.example .env # Populate .env with your GitHub OAuth credentials
-
Start Infrastructure
docker-compose up -d postgres redis
-
Run Backend
go mod tidy go run ./cmd/api
-
Run Frontend
cd frontend npm install npm run dev -
Visit
http://localhost:3000
We use Terraform for reproducible, click-to-deploy infrastructure.
-
Provision Infrastructure
cd terraform terraform init terraform apply -
Deploy Application
# 1. Login to ECR aws ecr get-login-password --region <region> | docker login --username AWS --password-stdin <account_id>.dkr.ecr.<region>.amazonaws.com # 2. Build & Push docker build -t codepair-dev . docker tag codepair-dev:latest <repo_uri>:latest docker push <repo_uri>:latest # 3. Update Service aws ecs update-service --cluster codepair-dev --service codepair-dev-api --force-new-deployment
This project is licensed under the MIT License.