Google Releases Jules Coding Agent to Take on OpenAI Codex
At the I/O 2025 event, Google released its Jules coding agent to all users worldwide. Jules is an autonomous coding agent that can read your codebase, understand your intent, and perform actions to get things done. It’s an asynchronous coding agent that connects to your existing repository and clones the codebase into a secure Google Cloud virtual machine (VM).
It means that the AI agent runs in the cloud, and not on your local machine. Just like OpenAI Codex, Google’s Jules AI agent can write tests, build new features, fix bugs, and update dependency versions. In addition, it can also provide audio changelogs.
You can assign tasks to the Jules coding agent, and it works in the background. When the task is completed, it shows its reasoning, plan, and a diff of the changes made. Google says your private codebase is not used for training AI models, and the data remains isolated in the secure environment.
Google has also added parallel execution to Jules, which means you can assign multiple tasks simultaneously. Moreover, you can connect your GitHub repo to understand the full context of your existing project. Finally, Google says since Jules is in public beta, it’s available free of charge, however, some usage limits do apply. You can access Jules via jules.google.com.
Source link
