GitHub Shows Off Copilot Workspace for Building Software from Scratch With Generative AI

GitHub has unveiled a new developer environment called GitHub Copilot Workspace that leverages generative AI to design and build software from start to finish. Copilot Workspace follows the success of GitHub Copilot generative AI pair programmer introduced in 2022.

Copilot Workspace

The GitHub Copilot Workspace is designed as a developer-native environment for coming up with ideas and going through all the planning, building, and testing of code using natural language. Instead of a single generative AI assistant, Copilot Workspace employs multiple Copilot agents that assist developers throughout the software creation process, from conception to completion. By integrating these tools, GitHub aims to reduce the barriers to entry for software development, making it more accessible for a broader range of individuals, including those with less 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,” GitHub CEO Thomas Dohmke explained in a blog post. “With Copilot Workspace we will empower more experienced developers to operate as systems thinkers, and materially lower the barrier of entry for who can build software.”

The platform takes a task-oriented approach where developers can start a new project directly from a GitHub Issue, Pull Request, or Repository. This is to help developers overcome the complexities of the initial steps of program creation. By providing AI assistance from the beginning, GitHub Copilot Workspace helps developers overcome initial hurdles and quickly move into the planning and coding phases.

Once a project is initiated, GitHub Copilot Workspace outlines a comprehensive plan for the developer, detailing the steps required to achieve the desired outcome. This plan is fully editable, allowing developers to customize and change their ideasGitHib made a point of emphasizing that it’s trying to augment, not replace, human creativity by automating routine tasks and providing suggestions, but how much that matters to developers isn’t clear.

Copilot Workspace isn’t the first improvement to GitHub’s toolkit since its debut. GitHub Copilot has since expanded to include coding by voice, an enterprise tier, and more proactive suggestions. GitHub Copilot Chat then came along, allowing users to seek advice by asking the AI assistant in plain language to explain coding concepts, translate between programming languages, and other valuable features. All of these improvements are visible in the Copilot Workspace. The generative AI programming assistant has proven popular, with Copilot making programmers likely to code faster and feel “more fulfilled with their job,” according to a 2022 survey of over 2,000 programmers.

Not that GitHub has the generative AI coding space all to itself. There’s an enormous amount of competition from tech giants, including the recently released Google Gemini Code Assist, Amazon’s CodeWhisperer, IBM’s watsonx Code Assistant, Meta’s Code Llama, and the open-source Starcoder 2 from Hugging Face. That’s before considering startups in the field attracting plenty of investment. For instance, generative AI coding assistance startup Tabnine scored a $25 million round in November, while February saw Codeium raise $65 million and Magic Ai raise $117 million for a coding ‘coworker.’ Last week alone saw generative AI coding assistant startup Augment Inc. raise $25 million, while Cognition Labs pulled in $175 million and claimed a $2 billion valuation.

Follow @voicebotaiFollow @erichschwartz

GitHub Starts a Coding Conversation With New Copilot Chat Generative AI Chatbot

Google Showcases New Generative AI Coding Assistants to Take on GitHub Copilot

Magic AI Raises $117M for Generative AI Coding ‘Coworker’ Not ‘Copilot