After quite some months of building ChatGPT-powered applications, I have realized that I always follow the same routine before starting a new project.
Any guesses?
👉 Right, it is not any fancy morning routine to boost my productivity, but it does have two steps:
Setting a new OpenAI API key for the incoming project — I try to give each project its own key so that I can monitor how much I spend developing the projects, among other security concerns.
Writing a standardized yet flexible method to call ChatGPT through the OpenAI API — I set up the ChatGPT call in a function at the beginning of the program so that the LLM usage is transparent to the application.
We already gave a shot at how to create an API key in the article A Step-by-Step Guide to Getting Your API Key, so in this article, I would like to share with you my preferred implementation to call ChatGPT from my Python applications.
