Android Studio Supercharges Development with ‘Agentic AI’ Journeys and Agent Mode

Google has announced a significant AI upgrade to Android Studio, the primary integrated development environment (IDE) for Android app developers, during its Google I/O 2025 developer conference. The update incorporates the latest Gemini 2.5 Pro model and introduces groundbreaking “agentic AI” capabilities, including the new Journeys feature and the upcoming Agent Mode, designed to streamline and enhance the app development process.

Journeys: Natural Language Testing with Gemini

The Journeys feature leverages the power of Gemini to enable developers to test their apps using natural language. Developers can describe the actions and expected outcomes of user journeys within the app, and Gemini will automatically perform these tests. This functionality aims to simplify the testing process, eliminating the need for extensive coding and allowing developers to validate app functionality more efficiently.

Google emphasizes that Journeys is still experimental but holds the potential to accelerate the delivery of high-quality code while significantly reducing the time required for testing, validation, and issue reproduction. Tests can be executed on both physical and virtual Android devices, with results displayed directly within the IDE.

Agent Mode: AI-Powered Assistance for Complex Tasks

The forthcoming Agent Mode represents another significant advancement, providing developers with an autonomous AI feature, also powered by Gemini, to handle more complex, multi-stage development tasks. For example, if a developer needs to integrate a new API, Agent Mode can create an execution plan that includes adding necessary dependencies, modifying files, and resolving bugs automatically.

Additional AI-Powered Features

Android Studio is also set to receive other AI-driven enhancements, including an improved “crash insights” feature within the App Quality Insights panel. This feature, powered by Gemini, will assist in identifying the root causes of app crashes within the source code and suggest potential fixes. Furthermore, developers can now access experimental AI features through a new “Studio Labs” menu within Android Studio’s Settings, available in stable releases starting with the Narwhal release.

Android Studio Cloud and Version Upgrade Agent

The public preview of Android Studio Cloud, accessible through Firebase Studio, provides a cloud-based development environment. This service streams a Linux machine running Android Studio directly to a web browser, enabling Android development from any location with an internet connection. Additionally, a Version Upgrade Agent, integrated with Gemini, is planned to automate dependency upgrades.

Enhanced UI Development with Gemini

Gemini’s capabilities extend to UI development, enabling developers to automatically generate Jetpack Compose preview code, transform UI code within the Compose Preview environment using natural language prompts, and attach image files or project files as context in chats with Gemini. The new “Rules in Gemini” feature will also allow developers to configure preferred coding styles and output formats.

Enterprise-Ready AI with Gemini Code Assist

Google is also launching an enterprise-ready version of Android Studio with Gemini in Android Studio for businesses. This offering allows teams to deploy AI while ensuring data security through subscriptions to Gemini Code Assist in Standard or Enterprise editions.

Further Updates and Improvements

Additional updates include resizable previews in Compose Preview, navigation improvements, an embedded Android XR emulator, and upgrades to Backup and Restore, as well as Backup and Sync. Android’s Kotlin Multiplatform will also benefit from several enhancements.

Preparing for Android’s 16KB Page Sizes

Google is also assisting developers in preparing for Android’s transition to 16KB page sizes, a significant architectural change, by providing early warnings and tools for testing apps in the new environment.