GitHub is previewing a new tool that leverages artificial intelligence (AI) to make relevant suggestions as programmers write code.
The system, called GitHub Copilot, has been trained on billions of lines of code available in public repositories, including those on GitHub. Microsoft and GitHub developed Copilot together with OpenAI, an AI research startup that Microsoft has been investing in since 2019.
“GitHub Copilot draws context from the code you’re working on, suggesting whole lines or entire functions. It helps you quickly discover alternative ways to solve problems, write tests, and explore new APIs without having to tediously tailor a search for answers on the internet,” wrote Nat Friedman, GitHub CEO.
Friedman describes GitHub Copilot as an AI-powered pair programmer. A popular collaborative programming technique, pair programming involves two developers working together to write and review code on the same screen.
More than autocomplete
Friendman explains that GitHub Copilot reviews the code you’re writing and makes relevant suggestions as you type. You can cycle through the suggestions, accept, edit, or reject them.
Furthermore, as with any AI-powered solution, Copilot learns over time. So as you write more code, Copilot will be able to make more relevant suggestions tuned to your coding style.
GitHub Copilot launches today in technical preview for a limited number of testers for free. For now, developers can only plug it into Microsoft’s Visual Studio Code, though it might support other IDEs later on.
Also, while Copilot is designed to work with a whole gamut of languages and frameworks, the technical preview is particularly adept at Python, JavaScript, TypeScript, Ruby, and Go.