Home / Companies / GitHub / Blog / Post Details
Content Deep Dive

How to write better prompts for GitHub Copilot

Blog post from GitHub

Post Details
Company
Date Published
Author
Rizel Scarlett, Michelle Duke
Word Count
1,963
Language
English
Hacker News Points
-
Summary

Generative AI coding tools like GitHub Copilot are revolutionizing how developers manage coding tasks, from documentation to unit test generation, though they present a learning curve that can lead to frustration when outputs are not as expected. Understanding how these tools process information and adjusting communication methods is crucial for effective use, as seen in examples where prompt adjustments led to more accurate code generation. This guide, authored by developer advocates Rizel Scarlett and Michelle Duke, explores best practices for prompt crafting with GitHub Copilot, emphasizing the importance of setting clear goals, simplifying requests, and providing examples. The document also highlights the necessity of adhering to good coding practices and experimenting with prompts to achieve desired results. Developers are encouraged to critically assess AI-generated code and to view the current phase of generative AI technology as one of trial and error and collective learning.