Make Your Data Models

Creating a clean and accurate data model is the foundation of any data-driven application. With SenseQL, we make it easy to design your data models visually and flexibly define how your tables relate to one another.

A link to another page why a data model is necessary, hyperlinked* 

You can build your data models in two ways:

• Let AI build it for you — Save time and generate relationships automatically.
  
  

• Manually define relationships — For full control and precision.

Let’s walk you through both options.

1. Auto-generate with AI

When working with datasets that have many interrelated tables, manually creating relationships can be tedious. SenseQL offers a smart AI-assisted option that automatically detects and maps out relationships for you.

To use it:

1. Once your tables are connected and displayed, click “Let AI create it for you”.
   
   

2. SenseQL will analyze the table names, column keys, and data patterns to generate the most likely one-to-one, one-to-many, or many-to-many relationships.
   
   

3. Review the model visually on the canvas.
   
   

4. If it looks correct, click “Looks Good” to confirm and proceed.
   
   

Perfect for quick starts or exploring unknown schemas.

2. Define Relationships Manually

Prefer to have complete control over your data model? You can define table relationships yourself.

To get started:

1. From the Define Relationships screen, click “I'll create it myself”.
   
   

2. On the canvas, select “Add” under the Relationships section of any table.
   
   

3. A modal will open to let you define the relationship in detail.

Here’s how to fill it out:

• From: Select the source table (e.g., employees)
  
  

• Field: Choose the column from the source table (e.g., employee_id)
  
  

• To: Choose the destination table
  
  

• Field: Select the column in the destination table that connects back (e.g., reports_to)
  
  

• Type: Define the relationship type (e.g., One To Many, Many To One)

Once filled, click Submit, and the relationship will appear visually on the modeling canvas.

Use manual mode when accuracy is critical, or when your schema includes legacy naming conventions or complex join logic.

What’s Next?

Once your data model is complete and relationships are defined, click “Looks Good” to continue. You’ll move on to Add Context, where you’ll define business logic by uploading files, providing SQL question-answer pairs, or generating table definitions so the agent understands both structure and purpose

Ready to build smarter models with SenseQL?