ChatGPT Prompt Engineering for Developers

Original link: https://limboy.me/links/chatgpt-prompt-engineering-for-developers/

ChatGPT Prompt jointly produced by Ng Enda and OpenAI. If you just watch the video, Bilibili has moved it . Although the title is for Developers , the code is relatively simple and highly readable. Even if you don’t know much about programming, you can basically understand it.

The video is not long, about one and a half hours in total, and describes how to give full play to the ability of chatgpt from different dimensions and combined with a suitable prompt. For example, by expressing clearly, chatgpt can be asked to perform the following tasks:

Wrap the context with a special tag, and then tell chatgpt to parse the content in the tag

Give the results in a fixed format for easy use in programs

Step disassembly, such as disassembling a text about making coffee into clear execution steps.

Through One Shot (give an example), let it absorb the essence of this example, and then we will ask questions, and we can give answers in the style of this example.

In addition, text summarization (such as compressing the customer’s review of the product into a few short paragraphs); model reasoning (such as inferring the customer’s emotions based on the text and giving different responses); text translation; Text expansion (under the premise that the content is basically the same, different styles of text can be written for different groups of people); chat robots. Things that used to be done with the help of machine learning models in the professional field can now be done with a large language model combined with a prompt in place. There are more scenarios and better things can be done.


PS: If you think the video is too long to find, you can also read the text summary of this Chinese version .

This article is transferred from: https://limboy.me/links/chatgpt-prompt-engineering-for-developers/
This site is only for collection, and the copyright belongs to the original author.