> ## Documentation Index
> Fetch the complete documentation index at: https://futureagi.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Create Dynamic Column by API Call

> The **API Call** feature allows users to dynamically fetch and populate new dataset columns by integrating external APIs.

Users can configure API parameters, headers, request body, and concurrency settings to process each row and extract relevant data.

## **1. Select a Dataset**

Before configuring the API Call column, ensure you have a dataset loaded. If no dataset is available, follow the steps to **Add Dataset** on the Future AGI platform.

***

## **2. Accessing the API Call Interface**

To create a dynamic column using an API call, navigate to your dataset and click the **+ Add Columns** button in the top-right menu. Scroll down to the **Dynamic Columns** section and select **API Call** to open the configuration panel.

***

## **3. Configuring the API Call**

* **Name**: Provide a name for the new column that will store the retrieved API response.
* **Output Type**: Select the format of the expected API response. Options include:
  * **String** (default)
  * **Object**
  * **Array**
  * **Number**
* **API Endpoint**: Enter the URL of the external API to fetch data.
* **Request Type**: Choose the appropriate request method:
  * **GET**
  * **POST**
  * **PUT**, **DELETE**, **PATCH**
* **Adding API Parameters and Headers**
  * **Params**: Define key-value pairs to send in the request query parameters.
  * **Headers**: Add authentication tokens, content types, or any required headers for API access.
* **Defining the Request Body**
  * If using **POST, PUT, or PATCH** requests, enter the request payload in JSON format.
  * You can use **{`{}`}** syntax to reference dataset column
* **Concurrency**: Define how many rows should be processed simultaneously.

***

Click **Test** to verify API connectivity and data retrieval. If the test is successful, click **Create New Column** to finalise the setup. The system will populate the new column dynamically with values fetched from the API.

***

## **Best Practices for Using API Calls**

* **Ensure API reliability**: Use APIs with stable endpoints and appropriate rate limits.
* **Validate output type**: Match the API response type with the selected output type.
* **Optimise concurrency settings**: Adjust based on dataset size and API rate limits for efficiency.

***