Integrating Kiro with Amazon SageMaker Unified Studio allows developers to utilize AI-assisted tools while maintaining strict data governance. This setup facilitates natural language queries for data exploration and analysis within a secure environment.
By connecting Kiro, VS Code, or Cursor to a SageMaker Unified Studio Space, users can access cloud-based compute resources while adhering to project permissions and data governance protocols. SageMaker automatically generates steering files, such as AGENTS.md, which provide essential context for the AI assistant, enhancing its ability to assist in data-related tasks.
Getting Started with Kiro and SageMaker
Setting up the integration involves a few straightforward steps:
- Ensure you have the AWS Toolkit for Visual Studio Code installed.
- Connect Kiro to your SageMaker Space through a secure SSH tunnel.
- Verify that your instance has at least 8 GiB of memory to enable Remote Access.
Once connected, the IDE will have full access to the file system, compute resources, and data services within the SageMaker Space.
Exploring Data with Kiro
With Kiro linked to SageMaker, users can begin exploring data using natural language prompts. For example, querying the sagemaker_sample_db.churn dataset allows Kiro to automatically retrieve and display the table schema and data samples.
This exploration helps Kiro understand the dataset's structure, enabling it to make informed decisions when running analyses. A best practice is to always start with data exploration before proceeding to code generation, as this reduces errors and enhances the accuracy of outputs.
Configuring Model Context Protocol (MCP) Servers
To effectively utilize Kiro's capabilities, configuring MCP servers is essential. These servers extend the functionality of Kiro, allowing it to query catalogs and run SQL commands. Users can prompt Kiro to discover and configure these servers, streamlining the setup process.
Utilizing Jupyter Notebooks
Kiro supports Jupyter notebooks, providing users with the same language and connection selectors available in SageMaker JupyterLab. Users can create notebooks directly within Kiro and run PySpark sessions as needed, making it easy to develop and test analytics workflows.
Key Takeaways
This integration between Kiro and SageMaker Unified Studio combines the strengths of AI-assisted coding with robust data governance. By following the outlined setup, users can enhance their data development processes while ensuring compliance with organizational standards.
Next Steps
To begin using this integration, users should download Kiro, install the AWS Toolkit for Visual Studio Code, and refer to the Amazon SageMaker Unified Studio documentation for detailed setup instructions.
