Overview

​

Introduction

Models are designed to define and organize data, simplifying the process of querying from various sources.

This guide outlines the process of creating a model, from selecting a data source to defining the model using various methods such as SQL queries, table selectors, or dbt models.

​

Understanding the Model Creation Process

Creating a model in AI Squared involves a series of steps designed to streamline the organization of your data for efficient querying. This overview will guide you through each step of the process.

​

Step 1: Navigate to the Models Section

To start defining a model:

1. Access the AI Squared dashboard.
2. Look for the Define menu on the sidebar and click on the Models section.

​

Step 2: Add a New Model

Once you log in to the AI Squared platform, you can access the Models section to create, manage, and monitor your models.

1. Click on the Add Model button to initiate the model creation process.

2. Select SQL Query, Table Selector, or dbt Model as the method to define your model.

​

Step 3: Select a Data Source

1. Choose from the list of existing connected data warehouse sources. This source will be the foundation for your model.

​

Step 4: Select a Modeling Method

Based on your requirements, select one of the following modeling methods:

1. SQL Query: Define your model directly using an SQL query.
2. Table Selector: For a straightforward, visual approach to model building.
3. dbt Model: Ideal for advanced data transformation, leveraging the dbt framework.

​

Step 5: Define Your Model

If you selected the SQL Query method:

1. Write your SQL query in the provided field.
2. Use the Run Query option to preview the results and ensure accuracy.

​

Step 6: Finalize Your Model

Complete the model setup by:

1. Adding a name and a brief description for your model. This helps in identifying and managing your models within AI Squared.

​

Unique Primary Key Requirement

• Essential Configuration: Every model in AI Squared must be configured with a unique primary key. This primary key is pivotal for uniquely identifying each record in your dataset.

• Configuring the Primary Key:

  • During the final step of model creation, select a column that holds unique values from your dataset.
  Ensuring the uniqueness of this primary key is crucial for the integrity and accuracy of data synchronization.

• Importance of a Unique Key:

  • A unique primary key is essential for effectively managing data synchronization.
  • It enables the system to track and sync only the new or updated data to the designated destinations, ensuring data consistency and reducing redundancy.

After completing these steps, your model will be set up and ready to use.