Google Tries Its Hand at Dev-Focused AI with 'Project IDX'

Google is leveraging its foundational AI model for an experiment called "Project IDX," which aims to helps developers create full-stack web and multi-platform applications in its cloud computing platform.

Designed to make it easier to build, manage and deploy full-stack web and multiplatform applications with popular frameworks and languages, Project IDX is a browser-based "development experience" built on Google Cloud. It is powered by Codey, a foundational AI model trained on code and built on PaLM 2, a machine learning large language model (LLM) developed by Google AI.

Although the experiment might be viewed as a challenger to Microsoft's VS Code for the Web and the similar Codespaces offering from Microsoft-owned GitHub (also based on VS Code), it actually uses "pure" open source tech similar to that used for Microsoft's open source-based VS Code app, called Visual Studio Code - Open Source ("Code - OSS").

"Project IDX lets you import your existing projects from GitHub so you can pick up right where you left off," Google said in an Aug. 8 post. "You can also create new projects, with pre-baked templates for popular frameworks, including Angular, Flutter, Next.js, React, Svelte, Vue, and languages such as JavaScript, Dart, and (coming soon) Python, Go, and more. We're also actively working to add first-class support for more project types and frameworks."

Google noted that in these early days, Project IDX already has smart code completion, an assistive chatbot and contextual code actions like "add comments" and "explain this code," with other functionality in the works.

That functionality is enabled by hooking into code models that Google calls Vertex AI Codey APIs, which include the following:

  • The code generation API - Generates code based on a natural language description of the desired code. For example, it can generate a unit test for a function. The code generation API supports the code-bison model. For more information about the code-bison model, see Code generation model parameters.
  • The code chat API - Can power a chatbot that assists with code-related questions. For example, you can use it for help debugging code. The code chat API supports the codechat-bison model. For more information about the codechat-bison model, see Code chat model parameters.
  • The code completion API - Provides code autocompletion suggestions as you write code. The API uses the context of the code you're writing to make its suggestions. The code completion API supports the code-gecko model. Use the code-gecko model to help improve the speed and accuracy of writing code. For more information about the code-gecko model, see Code completion model parameters.

Best practices for using the APIs include:

  • We recommend that a human is involved when the Codey APIs are used. Outputs of solutions created with the Codey APIs should be comprehensively tested before the solutions are used by customers in production.
  • Code generated by the Codey APIs is not intended or designed to be a replacement for code development.
  • We recommend that you don't use the Codey APIs to implement solutions for sensitive industries, such as cybersecurity and hacking prevention.

Project IDX taps into another Google Cloud service, Firebase Hosting for production deployment or generating a shareable preview of a web app.
Interested developers can apply to use Project IDX via a waitlist accessible from the site.

About the Author

David Ramel is an editor and writer for Converge360.