Real Time AI-Chat using Django and React

✅ Asynchronous chat (Channels, WebSocket)
✅ AI/LLM API consuming (OpenAI)
✅ Chat memory using Redis layer
✅ Production: Docker, Railway

For development or testing purposes

Add a .env file with the necessary variables to the root/project directory

OPENAI_API_KEY=<api-key>
SECRET_KEY=<secret-key>
DEBUG=True
ALLOWED_HOSTS=localhost,127.0.0.1

The app uses OpenAI API-key, but any provider's API should work (read the provider's documentation)

https://platform.openai.com/docs/quickstart?api-mode=chat

If Python is installed, a secret key can be generated with this command

Set to 32 [nbytes] for a 64 long key.

python -c "import secrets;print(secrets.token_hex(32))"

Running application as Docker image

docker build .

docker-compose -f docker-compose-dev.yml up --build

Removing containers:

docker-compose -f docker-compose-dev.yml down

Run the app in the web browser

http://localhost:8000

On Production

This app can be deployed to Railway

https://railway.com/

Key steps

• Add new project from GitHub
• Set domain on the port 8080
• Set the same variables except DEBUG in the Settings panel but change ALLOWED_HOSTS to the new domain
• Use Start Command: daphne -b 0.0.0.0 -p 8080 config.asgi:application
• Grab docker-compose-railway.yml from your file browser and drop on project canvas