On August 8, Google unveiled what it calls “Project IDX,” an experimental initiative aimed at moving developers’ entire cross-platform application development workflow to the cloud.
Generative artificial intelligence can make writing code more productive. With the IDX project, Google intends to explore how its AI innovations, including models Cody AND PaLM 2 powered by Studio Bot in Android Studio, Duet in Google Cloud, can help developers not only write code faster, but also write better code.
Project IDX is a web-based development environment (IDE) built on Google Cloud and powered by Codey, a core AI model trained on code and built on PaLM 2. It aims to facilitate the creation, management and deployment of complete web and cross-platform applications, with popular frameworks and languages. Google states that Project IDX is also built OSS codeMicrosoft’s GitHub repository for Visual Studio code development, adding ” so it should look familiar no matter what you’re building”.
Optimize applications using artificial intelligence and the cloud
Currently, Project IDX has, thanks to Codey, intelligent code completion, chatbot support and contextual code actions such as “add comments” and “explain this code”.
Developers will benefit from a comprehensive, high-performance development environment, no matter where they are or what device they’re using. Each Project IDX workspace has all the capabilities of a Linux-based virtual machine, combined with universal access provided by the cloud, in a nearby data center.
Google states:
“At the heart of the IDX project is our belief that you should be able to develop from anywhere, on any device, with full fidelity to local development “.
It is therefore possible to create projects with frameworks such as Angular, Flutter, Next.js, React, Svelte, Vue and programming languages such as JavaScript or Dart, as well as Python and Go. Developers can also import their existing projects from GitHub.
For developers to optimize their cross-platform apps, IDX Project allows them to preview them with an integrated web preview and soon test them with an Android emulator and an iOS simulator, available directly in the browser. Integration with Firebase Hosting allows them to set up a shared preview of web applications or put them into production.
Google researchers say they are exploring other ways to simplify and speed up the entire workflow, such as richer collaboration tools, environment synchronization, pairwise debugging, code review… and more personalized/contextual AI.
To that end, they are looking for feedback from developers on this first preview and are inviting them to register waiting list.
References: Google Blog
Authors: Bre Arder, Head of UX Research, Kirupa Chinnathambi, Product Manager, Ashwin Raghav Mohan Ganesh, Engineering Lead, Erin Kidwell, Director of Engineering, and Roman Nurik, Design Lead