Microsoft Corp.’s GitHub said today it’s making one of the main features of its enhanced copilot tool available to developers in limited beta test mode.

Github Copilot Chat is described as the “centerpiece feature” of Github Copilot X, an enhanced version of the popular Copilot coding assistant that debuted in mid-2021. Copilot X was announced in March, when the company said it will be embedded into the interface of Microsoft’s VS Code and Visual Studio code editing applications.

The beta availability of Copilot Chat is the first step in bringing GitHub Copilot X to enterprises, with an aim to transform the original Copilot into a “context-aware” conversational assist. According to the company, it lives right inside the integrated development environment, where developers create their code.

Copilot Chat is powered by generative artificial intelligence technology. It will enable developers to execute some of the most complex tasks involved in coding with simple prompts, speeding up the process of building new applications and debugging existing code.

GitHub explained that Copilot Chat is uniquely able to recognize the code that a developer has typed, as well as any error messages that are shown. It provides personalized assistance to every developer, with real-time guidance suggesting best practices, tips and solutions for each specific coding challenge the developer faces.

As well as providing advice on the next lines or snippets of code, it can automate code fixes by suggesting remediations, helping reduce the number of vulnerabilities found during security scans. Further, it can analyze entire chunks of code to explain in simple terms what that section is supposed to do. Finally, it can help with troubleshooting by identifying issues that may be causing problems with developer’s code.

In a blog post, GitHub Vice President of Product Mario Rodriguez said Copilot Chat is all about improving developer velocity. These days, developers spend increasing amounts of time doing noncoding tasks, he explained, such as searching through documentation to figure out how to connect to an application programming interface, use a new database or try to understand what a colleague was trying to accomplish.

Rodriguez cites Stack Overflow’s most recent developer survey, which found that 63% of developers spend between 30 minutes and two hours per day looking for answers and solutions. Of course, this means they’re spending less time on more productive work, such as creating new code and innovating solutions to problems.

“We want to help developers spend their time on what matters most: building what’s next,” Rodriguez explained. “This work started with GitHub Copilot offering code suggestions right in the IDE – but now with Copilot Chat, developers can not only get code suggestions, but ask questions, get explanations, offer prompts for code and more. That means they’re spending more time in the IDE and in the flow.”

Holger Mueller, an analyst with Constellation Research Inc., said improving developer productivity is a key focus for enterprises because there simply aren’t enough developers in the world to go around. “Most governments believe increasing developer velocity is critical for economic growth, and there’s a lot of hope for generative AI, which has already delivered the biggest productivity gains since the introduction of the PC itself,” Mueller enthused. “GitHub is fast becoming a leader here, integrating the latest advances in generative AI inside the IDE to help developers write superior code and debug it faster. Hopefully, this will help developers become more productive and ensure they spend far fewer late nights in the office eating pizza!”

The limited beta of Copilot Chat can be accessed by administrators now by navigating to the Copilot settings, clicking on the “Policies” tab and configuring the settings from within there.

Rodriguez said GitHub believes Copilot Chat will ultimately swing open the doors to a new age of code development, where natural language powers the entire experience for developers to make them happier and more productive.

Images: GitHub