Google LLC recently updated its Code Assist tool to enhance developer productivity in integrated development environments (IDEs) such as JetBrains and Visual Studio Code.

In a blog post, Google Product Management Director Ryan Salva and Group Product Manager Prithpal Bhogill announced that Code Assist will now utilize the Gemini 2.0 model, Google's most advanced large language model to date. This new model supports longer context windows and can handle larger codebases than before.

Additionally, Google plans to introduce several new Gemini Code Assist tools, which are currently available in a private preview. These new tools will integrate with data sources such as GitHub, GitLab, Google Docs, Atlassian, Sentry.io, and Snyk, allowing developers to request assistance from Code Assist directly within their IDE. Previously, Code Assist was limited to connecting only to the IDE.

With this update, Code Assist can now pull data in real-time and access information from third-party applications. Currently, access to the new version of Code Assist is restricted to Google Cloud partners.

Salva and Bhogill emphasized that the goal of the update is to provide more context for developers without disrupting their workflow.

Code Assist, formerly known as Duet AI, was relaunched for enterprise users last October. It was developed in response to the growing demand for AI-driven coding tools like GitHub Copilot and includes enhanced security and legal indemnification features. The previous update also improved code transformation capabilities and provided customized code suggestions based on customers' private code repositories.

Despite concerns about security, copyright, and reliability, developers have shown significant interest in AI programming assistants. A recent survey by GitHub revealed that the majority of developers have adopted some form of AI tools, with GitHub Copilot alone boasting over 1.8 million paying users.