Customize Amazon Q Developer in Amazon SageMaker Studio applications - Amazon SageMaker AI
Customize in JupyterLabCustomize in Code Editor
Services or capabilities described in Amazon Web Services documentation might vary by Region. To see the differences applicable to the China Regions, see Getting Started with Amazon Web Services in China (PDF).

Customize Amazon Q Developer in Amazon SageMaker Studio applications

You can customize Amazon Q Developer in the JupyterLab and Code Editor applications in Amazon SageMaker Studio. When you customize Q Developer, it provides suggestions and answers based on examples from your codebase. If you use Amazon Q Developer Pro, you can load any customizations that you've created with that service.

Customize in JupyterLab

In JupyterLab, you can load any customizations that you've created with Amazon Q Developer Pro. Or, in your JupyterLab space, you can customize Q Developer locally with files that you upload to the space.

To use customizations that you've created in Amazon Q Developer Pro

When you load a customization, Q Developer provides suggestions based on the codebase that you used to create the customization. Also, when you use the chat in the Amazon Q panel, you interact with your customization.

For more information about setting up customizations, see Customizing suggestions in the Amazon Q Developer User Guide.

To load your customization

Open your JupyterLab space and complete the following steps.

  1. In the status bar at the bottom of JupyterLab, choose Amazon Q. A menu opens.

  2. In the menu, choose Other Features. The Amazon Q Features tab opens in the main work area.

  3. In the Amazon Q Features tab, under Select Customization, choose your Q Developer customization.

  4. Interact with your customization in either of the following ways:

    • Create a notebook, and write code in it. As you do, Q Developer automatically provides tailored inline suggestions based on your customization.

    • Chat with Q Developer in the Amazon Q panel by following these steps:

      1. In the left sidebar in JupyterLab, choose the Jupyter AI Chat icon. The Amazon Q panel opens.

      2. Use the Ask Amazon Q chat box to interact with your customization.

To customize Amazon Q Developer with files in your JupyterLab space

In JupyterLab, you can customize Q Developer with files that you upload to your space. Then, in the chat in the Amazon Q panel, you can use a command to ask Q Developer about those files.

When you customize Q Developer with files in your space, the customization exists only in your space. You can't load the customization elsewhere, such as in other spaces or in the Amazon Q Developer console.

You can customize Q Developer with files in JupyterLab if you use either Amazon Q Developer Pro or Amazon Q Developer at the Free tier.

To customize with your files

Open your JupyterLab space and complete the following steps.

  1. Check whether your space is configured with the required embedding model. You can customize Q Developer in JupyterLab only if you use the default embedding model, which is CodeSage :: codesage-small. To check, do the following:

    1. In the left sidebar in JupyterLab, choose the Jupyter AI Chat icon. The Amazon Q panel opens.

    2. Choose the settings icon in the upper-right corner of the panel.

    3. For Embedding model, if necessary, choose CodeSage :: codesage-small, and choose Save Changes.

    4. In the upper-right corner of the panel, choose the back icon.

  2. To upload files that you want to customize Q Developer with, in the File Browser panel, choose the Upload Files icon.

  3. After you upload your files, in the Ask Amazon Q chat box, type /learn file path/. Replace file path/ with the path to your files in your JupyterLab space. When Amazon Q finishes processing your files, it confirms with a chat message in the Amazon Q panel.

  4. To ask Q Developer a question about your files, type /ask in the chat box, and follow the command with your question. Amazon Q generates an answer based on your files, and it responds in the chat.

For more information about the /learn and /ask commands, such as their options and supported arguments, see Learning about local data in the Jupyter AI user documentation. That page explains how to use the commands with the Jupyternaut AI chatbot. JupyterLab in Amazon SageMaker Studio supports the same command syntax.

Customize in Code Editor

If you've created a customization in Amazon Q Developer Pro, you can load it in Code Editor. Then, when Q Developer provides suggestions for your code, it bases them on the codebase that you used to create the customization. Also, when you use the chat in the Amazon Q: Chat panel, you interact with your customization.

To use customizations that you've created in Amazon Q Developer Pro

Open your Code Editor space and complete the following steps.

  1. In the Code Editor menu, choose View, and choose Command Pallette.

  2. In the command pallet, begin typing >Amazon Q: Select Customization, and choose that option in the filtered list of commands when it appears. The command pallet shows your Q Developer customizations.

  3. Choose your customization.

  4. Interact with your customization in either of the following ways:

    • Create a Python file or a Jupyter notebook, and write code in it. As you do, Q Developer automatically provides tailored inline suggestions based on your customization.

    • Chat with Q Developer in the Amazon Q panel by following these steps:

      1. In the left sidebar in Code Editor, choose the Amazon Q icon. The Amazon Q: Chat panel opens.

      2. Use the chat box to interact with your customization.

For more information about the capabilities of Q Developer, see Using Amazon Q Developer in the IDE in the Amazon Q Developer User Guide.