# Quickstart Guide

To start using the context builder, let's build a simple application. In this example, we will:

* Add a data source in the data settings.
* Create and refine a Conversational Retrieval QA to answer questions.
* Debug the application, have a conversation with AI, and review the returned results. 

Let's go step by step:

## #Create a new Dataset

After creating a data source, you will see an interface with 2 main configurations: 

Dataset name: Give your data source a name that will be used for future references. 

Document loaders: We allow you to seamlessly import data from any source. Our data loaders perform the following operations:

* Load data from the source.
* Convert data into text or arrays.
* Split the data into smaller segments (with content overlapping).
* Return a list of data segments.

<figure><img src="/files/Od7N29rHCOvCNlvtrsjF" alt=""><figcaption></figcaption></figure>

## #Create a new application

Let's start by creating a new application in "my app." You will see an interface with 3 pieces of information:&#x20;

* App Name: Give your application a name.&#x20;
* Short description: Provide a brief description of this application's features.&#x20;
* Image: Application icon or avatar.

<figure><img src="/files/9DnhDFCCC3FWKg7EvJth" alt=""><figcaption></figcaption></figure>

## #Explore the workflow interface

The "Add task" list provides two options:&#x20;

Tool: Offers pre-packaged, ready-to-use large model tools of different types.

First, let's add a tool type called Conversational Retrieval QA. All configuration options will be displayed in the right drawer. Only two settings are required to complete the process.

* In the "prompt" module, enter the model prompt.
* In the "data" module, add the data source you uploaded earlier.

<figure><img src="/files/guJDygF38KgEf6nF4UW6" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/Kgl6vWzQ2gk8dqNu0Wcu" alt=""><figcaption></figcaption></figure>

## #Debugging the app

Now that our workflow is configured, click the "enter debug" button to access the debugging interface. You can try having a conversation with it, asking some questions related to the data source, and see how it responds.

<figure><img src="/files/1xaA6gB5hBs5sRVrioBP" alt=""><figcaption></figcaption></figure>

## #What's next?

Now that we have built an app ready for production, we can continue maintaining and improving it. Some next steps to try are:

1. Increase debugging: Test your app with multiple inputs to evaluate its intelligence on a larger scale.
2. Share: Share the app with users who need it and gather their feedback.
3. Fine-tuning: Use the data collected from users to fine-tune your custom app and optimize its performance for your specific tasks.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://context-builder.gitbook.io/helpdocument/welcome-to-context-builder/quickstart-guide.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.