Generate SQL with the Data Agent

The SageMaker Data Agent is a conversational SQL assistant built into the query editor. Describe what you want to analyze in plain language, and the agent generates SQL queries based on your data catalog. The agent supports multi-turn conversations, so you can ask follow-up questions, request modifications, and get help fixing errors.

Open the Data Agent panel

1. In the query editor, choose the Chat with AI icon in the top-right action bar.

The agent panel opens on the right side of the editor with sample prompts and a chat input.

Ask a question and generate SQL

Type a natural language description of what you want to query. The agent analyzes your data catalog and generates SQL.

Example prompts:

• "What are the columns in the churn table?"

• "Which tables in the analytics schema have columns related to customer churn?"

• "Which customer groups have the highest churn?"

For complex analytical tasks, the agent proposes a step-by-step plan before generating SQL. You can review and approve the plan, or ask the agent to modify it.

Add generated SQL to your querybook

After the agent generates SQL, it automatically creates new cells in your querybook with the generated code. You can:

• Accept the generated SQL and run it in place.

• Reject the generated SQL to discard it.

• Accept and run to insert the SQL and run it immediately.

Multi-turn conversations

The Data Agent maintains context across messages within a conversation. You can build on previous queries by asking follow-up questions.

Fix errors with the Data Agent

When a query fails, the agent can analyze the error and suggest corrections.

1. Run a query that returns an error.

2. Choose Fix with AI on the failed cell.

3. The agent diagnoses the error and generates a corrected query.