Ada AI Coding Environment

A specialized AI coding assistant that helps with code analysis, debugging, and generation.

Features

• 🤖 AI-Powered Code Assistant: Uses Gemini 2.5 Flash Preview for intelligent code analysis
• 💬 Real-time Chat: Stream responses with markdown rendering
• 📝 Code Editor: Built-in CodeMirror editor with syntax highlighting
• 🔐 Secure Authentication: Firebase Auth with Google OAuth and Email/Password
• 💾 Chat History: Persistent storage in Firestore
• 🚀 Code Compilation: Direct integration with online compiler
• 📱 Mobile Responsive: Works seamlessly on desktop and mobile

Tech Stack

Frontend:

• Vanilla JavaScript (ES6+)
• Tailwind CSS
• CodeMirror (code editor)
• LZString (compression)
• Firebase SDK (Auth & Firestore)
• Marked.js (markdown rendering)

Backend:

• Python 3.x
• Flask
• Google Generative AI SDK (Gemini 2.5 Flash Preview)
• Firebase Admin SDK
• Flask-CORS

Database:

• Firebase Firestore (NoSQL)

Setup

Prerequisites

• Python 3.8+
• Node.js (for frontend development)
• Firebase account
• Google AI API key

Backend Setup

1. Clone the repository:

git clone https://github.com/notamitgamer/ada-web.git
cd ada-web

2. Install Python dependencies:

pip install -r requirements.txt

3. Create a .env file (use .env.example as template):

cp .env.example .env

4. Configure environment variables in .env:

GEMINI_API_KEY=your_actual_gemini_api_key
FIREBASE_CREDENTIALS={"type":"service_account",...}
PORT=5000

5. Run the backend:

python backend.py

Frontend Setup

The frontend is a static HTML application. Simply open index.html in a browser or serve it using a local server:

python -m http.server 5500

Then visit: http://localhost:5500

Firebase Configuration

Firestore Database Schema

users (collection)
  ├─ {uid} (document)
      ├─ email: string
      ├─ displayName: string
      ├─ photoURL: string
      ├─ createdAt: timestamp
      └─ chats (subcollection)
          └─ {chatUUID} (document)
              ├─ title: string
              ├─ createdAt: timestamp
              ├─ updatedAt: timestamp
              └─ messages: array[
                  {
                    role: "user" | "model",
                    content: string,
                    timestamp: timestamp
                  }
                ]

Firebase Service Account

1. Go to Firebase Console
2. Select project: ada-ai-aranag
3. Navigate to Project Settings > Service Accounts
4. Generate new private key
5. Use the downloaded JSON for FIREBASE_CREDENTIALS environment variable

API Endpoints

POST /api/chat

Streams AI responses for code queries.

Headers:

Authorization: Bearer <firebase_id_token>
Content-Type: application/json

Body:

{
  "message": "Why is this loop infinite?",
  "history": [],
  "codeContext": "while(i > 0) { ... }",
  "fileContext": "",
  "sessionId": "uuid-v4"
}

POST /api/generate-title

Generates a short title for chat sessions.

Headers:

Authorization: Bearer <firebase_id_token>
Content-Type: application/json

Body:

{
  "message": "First user message"
}

Deployment

Backend: Render.com

Quick Deploy

1. Create a new Web Service on Render.com

2. Connect your GitHub repository: notamitgamer/ada-web

3. Configure the service:

   • Name: ada-web-backend (or your preferred name)
   • Region: Choose closest to your users
   • Branch: main
   • Build Command: pip install -r requirements.txt
   • Start Command: gunicorn backend:app --workers 2 --threads 2 --timeout 120
   • Plan: Free (or paid for better performance)

4. Set Environment Variables:

   • GEMINI_API_KEY: Your Gemini API key (get from https://aistudio.google.com/app/apikey)
   • FIREBASE_CREDENTIALS: Full JSON string from Firebase service account
   • FLASK_DEBUG: False (for production)
   • PYTHON_VERSION: 3.11.0 (optional, auto-detected)

5. Deploy! Render will automatically deploy your backend.

Alternative: Using render.yaml

The repository includes a render.yaml file for automatic configuration. Simply connect your repo and Render will use this file for setup.

Health Check Endpoint

The backend provides a /health endpoint for monitoring:

• URL: https://your-app.onrender.com/health
• Response: {"status": "healthy", "timestamp": "..."}

Keep Backend Alive with UptimeRobot

Render's free tier spins down after 15 minutes of inactivity. Use UptimeRobot to keep it alive:

1. Create a free UptimeRobot account
2. Add New Monitor:
   • Monitor Type: HTTP(s)
   • Friendly Name: Ada Web Backend
   • URL: https://ada-web.onrender.com/health
   • Monitoring Interval: 5 minutes (free tier)
3. Save - UptimeRobot will ping your backend every 5 minutes to keep it alive!

Frontend (Static Hosting)

Deploy index.html to:

• GitHub Pages
• Netlify
• Vercel
• Any static hosting service

Update the API URL in index.html:

const API_URL = 'https://your-backend.onrender.com/api/chat';

Development

Running Tests

python test_backend.py

Code Style

• Python: Follow PEP 8
• JavaScript: ES6+ with consistent formatting
• Use meaningful variable names
• Comment complex logic

Security Features

✅ Firebase ID Token verification on backend ✅ CORS protection ✅ Environment variable protection ✅ Server-side authentication validation ✅ Secure API key storage

Contributing

1. Fork the repository
2. Create a feature branch
3. Commit your changes
4. Push to the branch
5. Open a Pull Request

License

MIT License - see LICENSE file for details

Author

Amit Dutta

• Website: amit.is-a.dev
• Project: Ada AI

Acknowledgments

• Google Gemini AI for powering the code assistant
• Firebase for authentication and database
• CodeMirror for the code editor
• The open-source community