Skip to content

Latest commit

 

History

History
149 lines (111 loc) · 4.38 KB

VERCEL_DEPLOYMENT.md

File metadata and controls

149 lines (111 loc) · 4.38 KB

Vercel Deployment Guide for TechEdu Hub Chatbot

This guide will help you deploy your Flask-based TechEdu Hub chatbot to Vercel for faster performance compared to Render.

Prerequisites

  1. Vercel Account: Sign up at vercel.com
  2. GitHub Repository: Your code should be in a GitHub repository
  3. API Keys: OpenRouter or OpenAI API key for chat functionality

Deployment Steps

1. Prepare Your Repository

Ensure your repository contains these files (already created):

  • vercel.json - Vercel configuration
  • api/index.py - Serverless function entry point
  • .vercelignore - Files to exclude from deployment
  • .env.vercel - Environment variables template
  • requirements.txt - Python dependencies

2. Connect to Vercel

  1. Go to vercel.com and sign in
  2. Click "New Project"
  3. Import your GitHub repository
  4. Vercel will automatically detect it as a Python project

3. Configure Environment Variables

In your Vercel project dashboard, go to Settings > Environment Variables and add:

Required Variables:

OPENROUTER_API_KEY=your_openrouter_api_key_here
SECRET_KEY=your_super_secret_key_here_change_in_production

Optional Variables:

FLASK_ENV=production
PYTHONPATH=.
DATABASE_URL=your_database_url_here
EMBED_MODEL=sentence-transformers/all-MiniLM-L6-v2
TOP_K=3
MAX_TOKENS=350
CHAT_MODEL=gpt-4o-mini

4. Deploy

  1. Click "Deploy" in Vercel
  2. Vercel will:
    • Install dependencies from requirements.txt
    • Build the serverless function
    • Deploy static files
    • Provide you with a live URL

5. Database Considerations

Option 1: SQLite (Default)

  • Works for development and small-scale production
  • Data is ephemeral (resets on each deployment)
  • No additional setup required

Option 2: External Database (Recommended for Production)

  • Use PostgreSQL, MySQL, or other cloud databases
  • Set DATABASE_URL environment variable
  • Examples: 
    # PostgreSQL
DATABASE_URL=postgresql://user:password@host:port/database

# MySQL
DATABASE_URL=mysql://user:password@host:port/database

6. Verify Deployment

After deployment:

  1. Visit your Vercel URL
  2. Test the chatbot functionality
  3. Check user registration and login
  4. Verify admin dashboard access
  5. Test FAQ search and AI responses

Key Differences from Render

Advantages:

  • Faster Cold Starts: Vercel's serverless functions start faster
  • Global CDN: Static files served from edge locations
  • Automatic Scaling: Scales to zero when not in use
  • Better Performance: Generally faster response times

Considerations:

  • Serverless Architecture: Each request is handled independently
  • Memory Limits: 1GB memory limit per function
  • Execution Time: 60-second timeout per request
  • Database: Consider external database for persistent data

Troubleshooting

Common Issues:

  1. Import Errors:

    • Ensure all dependencies are in requirements.txt
    • Check Python path configuration

  2. Database Connection:

    • Verify DATABASE_URL environment variable
    • Ensure database is accessible from Vercel

  3. API Key Issues:

    • Double-check OPENROUTER_API_KEY or OPENAI_API_KEY
    • Ensure keys have sufficient credits/permissions

  4. Static Files:

    • Templates and static files are handled automatically
    • Check vercel.json configuration if issues occur

Logs and Debugging:

  • View function logs in Vercel dashboard
  • Use Vercel CLI for local testing: vercel dev
  • Check browser developer tools for client-side errors

Performance Optimization

  1. Database Connection Pooling: Use connection pooling for external databases
  2. Caching: Implement caching for FAQ embeddings
  3. Async Operations: Use async/await for I/O operations where possible
  4. Memory Management: Monitor memory usage in Vercel dashboard

Monitoring

  • Vercel Analytics: Built-in performance monitoring
  • Function Logs: Real-time logging in dashboard
  • Error Tracking: Automatic error detection and alerts

Support

If you encounter issues:

  1. Check Vercel documentation: vercel.com/docs
  2. Review function logs in Vercel dashboard
  3. Test locally with vercel dev
  4. Check GitHub repository for configuration files

Note: This deployment maintains all features from the original Flask application while optimizing for Vercel's serverless environment.