Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.maia.ai/llms.txt

Use this file to discover all available pages before exploring further.

form your agentic data team. They help you to build, maintain, and operate your data pipelines. Whether you’re creating new workflows or optimizing existing ones, accelerate your work with powerful, natural language capabilities.

​
Video example

​
Enabling Maia AI Agents

are enabled at account level by default in , so you can start prompting them straight away. This means that by default, all users within your organization can ask for help building pipelines, creating custom connectors, and more. It also means that can sample data in any pipelines in your account. You can disable as a whole, or just their ability to sample data, at the account level. To do this, go to your account details, and use the Enable Maia and Enable sampling for Maia toggles to enable or disable these features. If are enabled at account level, users with the Member role cannot manage account-level settings for , but all other users can. For more information, see Account role permissions.

​
What can Maia AI Agents do?

The tasks that can perform for you include:

​
Pipeline development and design

  • Build orchestration pipelines to extract and load data from various sources into your data warehouse.
  • Create custom connectors for your orchestration pipelines.
  • Create transformation pipelines to transform, cleanse, aggregate, and enrich your data.
  • Design modular workflows by creating reusable, maintainable pipelines with variables and nested pipeline execution.
  • Validate and run pipelines from within the chat interface.
  • Manage pipeline and project variables for dynamic, reusable configurations.
  • Configure iterator and loop components for repeating tasks over datasets.
  • Disable and re-enable orchestration components without removing them from the pipeline.

​
Data exploration and visualization

  • Query your data warehouse to explore tables, views, and schemas.
  • Sample pipeline data to preview data at any stage of your pipeline without full execution.
  • Visualize data in charts directly in the chat interface to understand trends and patterns.
  • Search your projects to find specific content, such as configuration details, across all your files.
  • Validate pipeline components against your warehouse to catch configuration errors before execution.

​
Project and file management

  • Move and rename files to maintain clean folder structures.
  • Copy files, e.g. for use as templates for similar workflows.
  • Delete obsolete files to clean up unused pipelines and configurations.
  • Commit changes and push to your remote Git branch with auto-generated commit messages.
  • View uncommitted changes and diffs to review work before committing.

​
Pipeline analysis and optimization

  • Explain what your existing pipelines do, their components, and how they work.
  • Identify problems in pipeline configurations and suggest fixes.
  • Improve performance by recommending best practices for parallel execution, component selection, and workflow efficiency.

​
Consultation and best practices

  • Answer questions about your projects, components, and features of .
  • Advise on pipeline organization and data engineering.
  • Provide low-code expertise to help you leverage specialized components instead of custom SQL.
  • Create and maintain to-do lists for complex tasks so you can keep track of your progress.

​
Limitations

Working with is currently subject to the following limitations:
  • do not support scheduling or publishing pipelines.
  • Git integration is partial. can generate commit messages, commit changes, and push to a branch. However, they cannot perform advanced Git operations such as creating branches, resolving conflicts, or managing pull requests.