Skip to content

DEPLOY.md

Latest commit

 

History

History
96 lines (70 loc) · 2.55 KB

DEPLOY.md

File metadata and controls

96 lines (70 loc) · 2.55 KB

🚀 Agora Deployment Guide

Quick Deploy (5 minutes)

Frontend (Already Done ✅)

Live at: https://frontend-bqhveg6tl-federico-de-pontes-projects.vercel.app

Backend Deployment to Render

Step 1: Go to Render Dashboard

Step 2: Click "New +" → "Web Service"

Step 3: Connect your GitHub repository SCAILE-it/agora

Step 4: Configure the service:

  • Name: agora-backend
  • Region: Oregon (US West)
  • Branch: master
  • Root Directory: backend
  • Runtime: Python 3
  • Build Command: pip install -r requirements.txt
  • Start Command: uvicorn main:app --host 0.0.0.0 --port $PORT

Step 5: Add Environment Variable:

  • Key: OPENAI_API_KEY
  • Value: Your OpenAI API key (starts with sk-proj-...)

Step 6: Click "Create Web Service"

Step 7: Wait for deployment (~2-3 minutes)

Step 8: Copy your backend URL (will be something like https://agora-backend-xxxx.onrender.com)

Step 9: Update Vercel Frontend Environment Variables:

# In your terminal
cd ~/agora/frontend
vercel env add NEXT_PUBLIC_API_URL production
# Paste: https://your-backend-url.onrender.com

vercel env add NEXT_PUBLIC_WS_URL production
# Paste: wss://your-backend-url.onrender.com/chat/ws

# Redeploy frontend
vercel --prod

Alternative: Railway.app

  1. Go to Railway.app
  2. "New Project" → "Deploy from GitHub repo"
  3. Select agora repo
  4. Add environment variable OPENAI_API_KEY
  5. Railway will auto-detect Python and deploy

Alternative: Local Docker

cd ~/agora
docker compose up

Visit http://localhost:3000

Deployment Checklist

  • Frontend deployed to Vercel
  • Backend deployed to Render/Railway
  • Environment variables configured
  • Frontend updated with backend URL
  • Test the live application
  • Agents responding correctly
  • WebSocket streaming working

Troubleshooting

Frontend can't connect to backend

  • Check that backend is deployed and running
  • Verify CORS is allowing your frontend domain
  • Check environment variables in Vercel

WebSocket connection fails

  • Ensure using wss:// (not ws://) for production
  • Check backend logs for errors
  • Verify WebSocket endpoint /chat/ws is accessible

Agents not responding

  • Check OpenAI API key is set correctly
  • Check backend logs for OpenAI API errors
  • Verify you have OpenAI credits

Support