> For the complete documentation index, see [llms.txt](https://documentation.virtuedeep.tech/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://documentation.virtuedeep.tech/getting-started-with-virtue/getting-started-guide.md).

# Getting started guide

A quick start guide to working with our platform.

This guide describes how to get started with building an AI solution powered by Affective Computing by Virtue:

{% stepper %}
{% step %}

### Design your human-powered AI solution's architecture

First, determine how you want to leverage our framework's capabilities. You can do the following:

1. Identify your solution's domain and define its real-world applications.
2. Identify which [Affective Computing by Virtue services](https://documentation.virtuedeep.tech/affective-computing-by-virtue/affective-computing-by-virtue.md) you need.
3. Determine whether you need to build a solution from scratch or integrate an existing solution with Virtue's platform.

Learn more:

* [How to set up a solution](https://documentation.virtuedeep.tech/affective-computing-by-virtue/designing-solutions-powered-by-affective-computing/how-to-set-up-a-solution.md#building-a-blue-ocean-solution)
* [How to set up a solution](https://documentation.virtuedeep.tech/affective-computing-by-virtue/designing-solutions-powered-by-affective-computing/how-to-set-up-a-solution.md#building-a-domain-specific-solution)
* [How to set up a solution](https://documentation.virtuedeep.tech/affective-computing-by-virtue/designing-solutions-powered-by-affective-computing/how-to-set-up-a-solution.md#considerations-when-building-a-solution-from-scratch)
  {% endstep %}

{% step %}

### Connect a language model

Before your tenant can feed any content, it needs a language model connected. In the Portal, go to **Developers Hub → Third-party Connections → Language Models (LM)** and either connect your own provider — AWS Bedrock, for example — or select one of the open-source models offered. Check the add-on price before selecting an offered model.

{% hint style="warning" %}
This is a prerequisite, not an optional step. The feeding wizard will not let you proceed until a model is connected.
{% endhint %}

Learn more:

* [Configuring LLMs](https://documentation.virtuedeep.tech/guides/using-the-public-apis/configuring-llms.md)
  {% endstep %}

{% step %}

### Choose and declare your data inputs

Next, determine the data inputs your solution requires and plan how to collect these inputs from users.

{% hint style="success" %}
Affective Computing by Virtue supports inputs of the following types:

* **Physiological inputs**: Involuntary responses of the human body such as blood pressure, heart rate, sweat, breath, and more.
* **User motion inputs**: Data about physical movements, reactions, and behavior of users, such as speech, joystick movement, and more.
* **Environmental inputs**: Inputs based on the stimuli that surrounds users, such as location, time, weather, and more.
  {% endhint %}

After choosing inputs, declare them in your Affective Computing by Virtue project. You can do this when creating or editing the project.

Learn more:

* [Understanding data inputs](https://documentation.virtuedeep.tech/affective-computing-by-virtue/designing-solutions-powered-by-affective-computing/understanding-data-inputs.md)
  {% endstep %}

{% step %}

### Connect your front end to Sentio

After declaring data inputs, enable the data to flow into your solution.\
You can do this by connecting the front end of your app, device, or data source to Sentio, our infrastructural platform.

{% hint style="success" %}
All endpoints of Sentio APIs follow REST conventions and use standard HTTP methods.
{% endhint %}

1. (If applicable) Set up any connections required at the data sources' ends.\
   For example, this could mean enabling physical sensors that collect data from end users to send measurements over BLE.
2. Authenticate the front end of your data sources to use our APIs.
3. Enable your front end to pass inputs to Affective Computing by Virtue through the parameters of the `Interaction/Interact` API.

Learn more:

* [How to set up a solution](https://documentation.virtuedeep.tech/affective-computing-by-virtue/designing-solutions-powered-by-affective-computing/how-to-set-up-a-solution.md#considerations-when-building-a-solution-from-scratch)
* [Authenticating your data source to use our APIs](https://documentation.virtuedeep.tech/guides/using-the-public-apis/getting-started/authenticating-your-data-source-to-use-our-apis.md)
* [Perform an interaction](https://documentation.virtuedeep.tech/guides/using-the-public-apis/creating-managing-and-performing-interactions/perform-an-interaction.md)
  {% endstep %}

{% step %}

### Start building your solution

After successfully plugging in your solution to Virtue's platform, you can start building your solution based on your planned architecture and at your convenience.

You can choose one of the following approaches:

* [Implementing a direct integration](https://documentation.virtuedeep.tech/getting-started-with-virtue/implementing-a-direct-integration.md): Parameterize your project using our no-code SaaS GUI Portal.
* [Implementing an advanced integration](https://documentation.virtuedeep.tech/getting-started-with-virtue/implementing-an-advanced-integration.md): Parameterize your project through a low-code integration with our public APIs.

{% hint style="success" %}

#### Which solution integration approach should you choose?

* If you have low technical expertise and prefer to use a no-code, GUI-based platform for simple and quick solution design, you can use the Portal to implement a direct integration.
* As a developer, if you prefer to program your solution with fine-grained control or if your solution requires advanced customization, you can use our APIs to implement an advanced integration.
  {% endhint %}
  {% endstep %}
  {% endstepper %}
