This guide covers deploying FitTrack AI to production cloud platforms.

---

1. Create a free M0 cluster on MongoDB Atlas.
2. Create a Database User with read and write credentials.
3. In Network Access, add IP address 0.0.0.0/0 to allow connections from cloud hosting platforms.
4. Copy your Connection String URI:

   mongodb+srv://<username>:<password>@cluster0.mongodb.net/fittrack?retryWrites=true&w=majority
   

---

1. Create a new Web Service on Render.
2. Connect your GitHub repository.
3. Configure the service:
   • Root Directory: server
   • Environment: Node
   • Node Version: >=20.0.0
   • Build Command: npm install
   • Start Command: node server.js
4. Add Environment Variables:
   • NODE_ENV: production
   • PORT: 10000
   • MONGODB_URI: Your MongoDB Atlas Connection String
   • JWT_SECRET: A secure random secret (min 32 chars)
   • OPENROUTER_API_KEY: Your OpenRouter API Key
   • CLIENT_URL: https://ai-fitness-tracker1.vercel.app
5. Deploy service. Note your public backend URL (e.g. https://fittrack-api.onrender.com).

---

1. Sign in to Vercel and click Add New > Project.
2. Select AI-FitnessTracker1.
3. Configure Project Settings:
   • Root Directory: client
   • Framework Preset: Vite
   • Build Command: npm run build
   • Output Directory: dist
4. Add Environment Variable:
   • VITE_API_URL: https://fittrack-api.onrender.com (Legacy VITE_STRAPI_API_URL is also supported for backward compatibility).
5. The repository includes client/vercel.json for client-side single page app routing:

   { "rewrites": [{ "source": "/(.*)", "destination": "/" }] }

6. Click Deploy.

---

1. Verify that the GitHub Actions CI workflow (.github/workflows/ci.yml) completed cleanly.
2. Navigate to your deployed Vercel domain.
3. Sign up a new user account.
4. Verify that the dynamic goal rings render and the water tracker increments properly.
5. Inspect the 90-Day Calendar Heatmap on Food Log and Activity Log.
6. Test the AI Food Snap by uploading an image.
7. Open AI Assistant and chat with FitBot.