Meta launched Muse Code, an AI agent that helps with programming, planning changes, and validating large software projects
As Zuckerberg explained, when a task is large enough, the system distributes the work among several subagents that operate in parallel within isolated environments. This way, the main copy of the project is not modified during the process.
As an example, the executive indicated that during testing, the tool managed to develop six functions for a video game at the same time without generating conflicts among them.
Meta launched Muse Code, a new artificial intelligence designed to help developers work on large codebases and automate complex software engineering tasks.
The tool is already available in beta and aims to compete with other solutions in the market, offering assistance for planning changes, writing code and validating project results.
Muse Code is the new programming agent introduced by Meta. According to the company's CEO, Mark Zuckerberg, the tool can complete software engineering tasks on large code repositories.
Its functions include planning changes, generating code, and validating the results obtained during development.
Muse Code can be installed with a single command and operates on Muse Spark, the programming model that Meta had previously introduced.
To tackle large projects, the agent creates independent processes that work simultaneously on different parts of the code.
With the launch of Muse Code, Meta seeks to position itself against other AI-assisted programming platforms.
The company aims to compete with Codex from OpenAI, and Claude Code, developed by Anthropic.
Alexandr Wang, head of artificial intelligence at Meta Superintelligence Labs, stated in an interview with the Wall Street Journal that the company believes the tool can become a very competitive option, especially from a cost perspective.
The launch of Muse Code is part of Meta's strategy to expand its presence in the artificial intelligence market.
In June, the company also entered the enterprise AI segment with an agent focused on customer service and support tasks, thus expanding the use of this technology beyond its advertising business.
