Posted by Sandhya Mohan – Product Supervisor, Android Studio and Sarmad Hashmi – Software program Engineer, Labs
The Android workforce believes AI has the potential to revolutionize coding, drive unprecedented innovation and productiveness in software program improvement, and supercharge your improvement productiveness. AI code completion is a key a part of this effort inside Gemini in Android Studio.
Since launching in Might 2024, we have been arduous at work enhancing this characteristic to offer the absolute best expertise for all Android builders. On this put up, we wish to take you “beneath the hood” on how we achieved a 40% relative enhance in acceptance charge since launch, and share a few of our pleasure for the way we have now seen Android builders use this characteristic. We hope you will give it a attempt to tell us what you suppose.
An AI coding companion for each developer
Our imaginative and prescient for Gemini in Android Studio is to empower builders to construct top quality Android apps — making it simple for builders to shortly write appropriate code aligned with Android’s finest practices. Launched final yr, the primary model of Studio Bot offered a chat expertise the place builders might entry Android-specific steerage, powered by Google’s newest AI fashions. Builders are capable of ask Gemini in Android Studio to offer developer steerage, summarize technical documentation, and critique their Android code. However in all these instances the suggestions is reactive, responding to a consumer’s query.
AI code completion takes these capabilities a step additional by offering real-time suggestions as you’re employed as a developer, considering forward and suggesting the subsequent few strains of code that you’re prone to kind based mostly on the context from the encircling file and what was simply typed. You may consider AI Code Completion as a companion in your work — a coding companion ready to supply steerage whenever you want it.
This characteristic is especially nicely fitted to duties like defining enterprise logic, creating database schemas, making community requests, and even writing checks — duties which might be typically time-consuming and distract from constructing the core expertise on your app. Many builders have instructed us how a lot they benefit from the velocity AI completions brings to their app improvement workflow.
.gif)
Bringing extra clever code completion to Android improvement
Whereas we’re excited to see how AI Code Completions have improved builders’ workflows, we all know there’s nonetheless extra we will do to enhance developer productiveness. Growth of Gemini in Android Studio is an ongoing, large-scale collaborative effort by many groups throughout Google. Earlier this yr, we switched to Gemini 1.5 fashions and noticed a major enchancment within the high quality of code completions, leading to a 2x enhance in our developer productiveness metrics, together with total acceptance charge for solutions.
As soon as we began doing A/B check experiments to enhance AI code completion we discovered a number of enhancements round mannequin high quality, context, and heuristics. This total effort led to a 40% relative enhance in acceptance charge — how typically customers settle for the AI’s proposed code solutions — since we launched. Since then, we have been exploring a number of enhancements like:
- Retrieval augmentation: Along with your opt-in consent, we use the information and dependencies most related to your present coding context to reinforce the accuracy of solutions. That is simply step one and we’re persevering with to experiment with including much more context from the IDE as a part of every request.
- Filtering out low-confidence completions: Prioritize exhibiting top quality solutions the place they’re most related, and subsequently most certainly to be accepted. We do that by utilizing a mix of the chances returned by the mannequin and utilizing a classifier educated to determine high-quality completions based mostly on developer suggestions.
- Smarter post-processing: The LLM’s output for AI Code Completion is basically totally different from the output customers anticipate in a chat session. Responses must be tightly scoped with a purpose to shortly output helpful code, with out surrounding expository textual content. We apply extra heuristics on the mannequin output to make sure responses are concise and correct, in addition to ensuring that the generated code is legitimate throughout the context of the consumer’s codebase.
- Improved fashions: We use opt-in suggestions from Android Studio customers, corresponding to noting when a code suggestion is accepted or rejected, to adapt the code completion mannequin to their coding type and preferences over time. We often ship new fashions with greater high quality knowledge based mostly in your suggestions.
We’re additionally exploring metrics past acceptance charge to raised measure AI impression on developer velocity, corresponding to the share of whole code written by AI.
Attempt it out!
We’re rolling out these profitable experiments and others as shortly as potential.
If you have not tried AI code completions but, you may allow this characteristic by clicking on the 
Gemini button in your editor window and signing in to your Google account.
After doing so, navigate to Settings > Instruments > Gemini and choose “Allow AI-based inline code completions”.
As at all times, Google is dedicated to the accountable use of AI. Android Studio will not ship any of your supply code to servers with out your consent — which suggests you will must opt-in to allow Gemini’s developer help options in Android Studio. You may learn extra on Gemini in Android Studio’s dedication to privateness.
Attempt enabling AI Code Completions in your undertaking and inform us what you suppose on social media with #AndroidGeminiEra. We’re excited to see how these enhancements aid you construct wonderful apps!
This weblog put up is a part of our collection: AI on Android Highlight Week, the place we offer assets — weblog posts, movies, pattern code, and extra — all designed to to discover the most recent in AI and its potential for Android app improvement.
