AI-powered codebase understanding platform built for AWS AI for Bharat Hackathon.

Developers spend 60% of their time reading and understanding code. Onboarding to new codebases is slow and frustrating.

CodeBase Navigator uses AI (Groq API with Llama 3.1 8B Instant) to help developers understand any GitHub repository through natural conversation, visual diagrams, and smart code navigation.

• 🔍 GitHub Repository Integration - Analyze any public repository
• 💬 Natural Language Queries - Ask questions about code in plain English
• 🌲 Smart File Navigation - Interactive file tree with intelligent highlighting
• 📊 Code Visualization - Dependency graphs and architecture diagrams
• 🏗️ Architecture Overview - Auto-generated project summaries
• 🎨 Modern UI - Dark mode, responsive design, syntax highlighting

• Frontend: React + Vite + Tailwind CSS
• Backend: Node.js + Express
• AI: Groq API (Llama 3.1 8B Instant)
• Code Analysis: Custom AST parser
• Deployment: Vercel (Frontend + Backend)

• Node.js 18+
• Groq API key (Get one free here)

1. Clone the repository:

git clone https://github.com/lekhanpro/Codesense.git
cd Codesense

cd codebase-navigator

2. Install dependencies:
```bash
npm run install:all

3. Set up environment variables:

Create backend/.env:

GROQ_API_KEY=your_groq_api_key_here
GITHUB_TOKEN=your_github_token_optional
PORT=3001
FRONTEND_URL=http://localhost:5173

4. Start development servers:

npm run dev

Frontend: http://localhost:5173 Backend: http://localhost:3001

1. Enter a GitHub repository URL (e.g., https://github.com/facebook/react)
2. Wait for analysis to complete
3. Explore the file tree and ask questions like:
   • "How does authentication work?"
   • "Explain the main function"
   • "Show me the data flow"
   • "What design patterns are used?"

Try these pre-loaded examples:

• React Todo App
• Express REST API
• Python Flask Application

┌─────────────┐      ┌──────────────┐      ┌─────────────┐
│   React     │─────▶│   Express    │─────▶│  Groq AI    │
│  Frontend   │      │   Backend    │      │  (Llama 3.1)│
└─────────────┘      └──────────────┘      └─────────────┘
       │                     │
       │                     ▼
       │              ┌──────────────┐
       └─────────────▶│  GitHub API  │
                      └──────────────┘

cd frontend
vercel --prod

cd backend
vercel --prod

Update frontend/.env.production with your backend URL.

• ⚡ 78% faster developer onboarding
• 🎯 4.5x better code comprehension
• 💰 $13.85B TAM - 27.7M developers globally
• 🌍 Supports repositories up to 10,000 files

• Private repository support
• Export explanations as Markdown
• Multi-repo comparison
• Auto-generate onboarding docs
• IDE extensions (VSCode, IntelliJ)
• Team collaboration features

MIT License - see LICENSE file for details

Built with ❤️ for AWS AI for Bharat Hackathon 2026

• Groq for fast AI inference (Llama 3.1 8B Instant)
• GitHub for repository access
• AWS for hosting infrastructure