Travel GPT Backend

This is the backend API server for o~dyssey AI.

System Architecture

• AWS S3

  • Stores generated images.

• Supabase

  • Serverless PostgreSQL for data storage.

• OpenAI

  • GPT model
    • Generates suggested mood messages.
  • DALL·E model
    • Creates images based on mood messages.

• SerpAPI

  • Enables Google Lens image searches.

• Google APIs

  • Powers Google Map related features such as finding nearby spots.

• Sentry

  • Monitors the backend server's health and performance.

• Fly.io

  • Serverless platform for backend API server deployment.

Implementation

Requirements

• Python
• Pipenv

Tech Stack

• FastAPI
  • Gunicorn
    • WSGI HTTP server (supports multiple workers).
  • Uvicorn
    • ASGI web server (asynchronous).
    • [Running Uvicorn with Gunicorn]:

      Uvicorn includes a Gunicorn worker class allowing you to run ASGI applications, with all of Uvicorn's performance benefits, while also giving you Gunicorn's fully-featured process management.

Get Started

Environment Variables

• Set up the environment variables file .env by referencing .env.example.
• Run the following command or use a plugin like autoenv to load the variables:

  source .env

Python Packages

• Install the Python virtual environment management tool Pipenv:

  python -m pip install pipenv

• Install python packages with Pipenv

  python -m pipenv sync

Backend API Server

• Start the server by running the following command:

  python src/servers/server.py

• Test the API server:

  # test local server with curl
  curl -XGET 'http://APP_HOST:APP_PORT/healthz' -H 'Authorization: Basic APP_AUTH_TOKEN'
  
  # response
  {"message":"endpoint: /healthz, message: running"}