Prompt Engineering for Generative AI

Chapter 1. The Five Principles of Prompting

Prompt engineering is the process of discovering prompts that reliably yield useful or desired results.

A prompt is the input you provide, typically text, when interfacing with an AI model like ChatGPT or Midjourney. The prompt serves as a set of instructions the model uses to predict the desired response: text from large language models (LLMs) like ChatGPT, or images from diffusion models like Midjourney.

Here is a simple example of a prompt input for a product name generator (inspired by one of OpenAI’s examples), and the resulting output from ChatGPT.

Input:

Can I have a list of product names for a pair of shoes that
can fit any foot size?

Output:

Certainly! Here's a list of product names for a pair of shoes
that can fit any foot size:

UniFit SoleStrides
FlexiSize All-Fit Shoes
OmniFit FootFlex
...
AnyStep AdaptiWalkers
FlexiRange UniversalSteps
PerfectGlide MultiFit Shoes

This is already a remarkable response for a naive prompt, which feels like magic because we got here with very little effort. As the state-of-the-art models improve, the likelihood you will get good enough results on your first try goes up. For any throwaway interactions with an AI, where you don’t plan to do the same task again, the naive approach is all you need.

However, if you planned to put this prompt into production, you’d benefit from investing more work into getting it right. Mistakes cost you money in terms of the fees OpenAI charges based on the length ...