Microsoft Corp.’s GitHub today announced a new generative artificial intelligence-powered development environment called Copilot Workspace, which builds on the capabilities of its original AI-powered coding tool to help reduce the time it takes to get started on new projects.

Unveiled in a blog post authored by GitHub Chief Executive Thomas Dohmke, Copilot Workspace is designed to help developers by reducing the time they need to spend reading through an existing codebase to work out how to get started on a new project. They can use natural language to brainstorm ideas, plan, build, test and run their code, and do it faster than before.

The new offering is an extension of GitHub’s original Copilot tool, which was launched in beta back in 2021 and quickly became one of the most iconic early generative AI applications, helping developers by suggesting lines of code as they worked. GitHub Copilot was followed by Copilot Chat, which enables developers to test and debug their code using natural language.

Copilot Workspace, which was first teased last year, provides developers with end-to-end support for any project, and it is designed to help them at one of the most challenging moments – the very beginning, where many coders suffer from a kind of writer’s block.

As Dohmke explained, for most developers, one of the greater barriers is the start of any new project. “Think of how often you hit a wall in the first steps of a big project, feature request, or even bug report, simply because you don’t know how to get started,” he explained. “GitHub Copilot Workspace meets developers right at the origin: a GitHub Repository or a GitHub Issue. By leveraging Copilot agents as a second brain, developers will have AI assistance from the very beginning of an idea.”

As soon as Copilot Workspace understands what the developer is trying to accomplish, it will offer a step-by-step plan written in natural language to help them get there, based on its unparalleled knowledge of the codebase they’re working with. As the developer works, it will throw up helpful suggestions, reply to questions, fix bugs and more. It provides users with everything they need to carry out and validate their plan of action and test the code that results.

“This new task-centric experience leverages different Copilot-powered agents from start to finish, while giving developers full control over every step of the process,” Dohmke said.

Every step and code suggestion is editable, which means human developers maintain full control over the entire process, but the advantage is it reduces much of the manual work, so developers don’t have to build every single component themselves. Once the code has been completed, the user can then run it within Workspace, make tweaks in GitHub Codespace, and share the finished product via a link to other collaborators, who will be able to see exactly which parts of the code were written by Copilot’s AI agents.

The original GitHub Copilot made a splash when it was first released, and it has been updated multiple times since then, the most recent addition an enterprise plan.

But rather than just assist developers, GitHub now wants to transform the entire experience. “Copilot Workspace represents a radically new way of building software with natural language, and is expressly designed to deliver — not replace — developer creativity, faster and easier than ever before,” Dohmke said.

The basic idea is that by making code easier to create, professional developers can focus on the bigger picture instead of getting bogged down in hundreds of lines of code, Dohmke said. What’s more, it will aid every kind of developer – not only the professionals, but also beginners and hobbyists.

When it launches, GitHub says the platform will “democratize coding” for developers of all levels by “reducing boilerplate work.”

The company said GitHub Copilot Workspace is available now as a technical preview for waitlisted developers, and those who are interested in checking it out can sign up here. Once all of the kinks have been ironed out and the platform is ready for prime time, the company will integrate it into the existing GitHub Copilot offering.

Image: GitHub