Prompt Engineering Course by OpenAI — Inferring, Transforming, and Expanding with ChatGPT

Maximize ChatGPT’s Potential in your Custom Application

Image by Kelly Sikkema in Unsplash. Brand logos have been added manually.

OpenAI has released a free course on prompt engineering in partnership with the learning platform DeepLearning.AI. This short course taught by Isa Fulford and Andrew Ng provides a collection of best practices to use Large Language Models (LLM) such as ChatGPT as part of your custom application.

The course is supposed to be free for a limited time, which is why I decided to do a series of short articles on the take-home messages expanding the information provided in the course.

In this follow-up article, we will down deep into three tasks that ChatGPT is especially good at: interring, transforming, and expanding. In the course, the instructors encourage developers to use ChatGPT in their applications when any of these tasks is required due to its high capability at performing those tasks.

⚠️ The examples in this article are mostly taken from the original course. The examples can be found in this Jupyter Notebook.

ChatGPT API

Here is the basic code structure we will use to call the ChatGPT API in the following examples: