under construction

Purpose

• Make it easy to create and deploy applications that use AI
• It is more fun and rewarding to have something actually finished, deployed, and useable as a starting point that you can then build on top of.
• You will better discover what you actually need by building MVPs in this way.

Limitations

• This guide and code is not meant for large scale production. It is meant to be a starting point for people who want to learn how to use AI in their applications. Creating an application for yourself and or a few people.
• It is optimized to be quick and easy to start, deploy, and learn from.

How to use

• Get an open AI key. set a low budget so you don't get surprised by the bill. You can always increase it.

• fork this repo

• clone your fork

• create a virtual environment

• install the requirements using pip install -r requirements.txt

• run the app using streamlit run app.py

Using with codespaces (No local setup required. Code in your browser)

• see https://github.com/features/codespaces for detailed instructions
• Fork this repo
• go to settings -> codespaces -> click set up prebuild
• in the repo, under code, click "start codespace"
• to launch from command line run python -m streamlit run app.py

customizing the application

• see RPG app for an example of how to customize the application

features

• Estimates cost of API requests so you don't get surprised by the bill
• Includes conversation history and allows you to change how much if any to use

Deployment

Put your application on the web so you can easily access it

suggestions

• Use chatgpt and or github copilot to help you write and troublelshoot your code
• when your code gets to be long, consider some refactorings:
  • extract methods
  • separate display logic from business logic
• Use gpt3.5-turbo when you can instead of gpt-4. It is much cheaper.

Resources

Prompt engineering

Other APIs

Demo of this app

demo of some apps I have created that use streamlit

Going further(Wait until you actually need this)