HomeGuidesAPI ReferenceRelease notes
Log In
Guides

Build an application

Rapidly build, customize, and integrate enterprise-grade AI capabilities into any application.

Suggest Edits

Before you start..

SWE offers you the ability to create a Generative AI application with multiple capabilities. Here are some important tips to get you started:

  • Studio Area: Upon creation, you'll be directed to our studio area. In this space, any changes you make are live and immediately reflected in the chat playground without needing to publish the application or changes.
  • Chat Playground: After connecting a model, you can start using the chat functionality directly, without needing to deploy the app or save changes. All modifications in the Studio area automatically affect the chat playground.
  • Preview Mode: Preview mode allows you to see and experience the chat exactly as your users will.
  • Important Notes:
    • When you're in the studio area, your changes are saved temporarily. However, if you leave the Studio (navigate to another page or refresh it), your changes will be lost. To preserve your modifications, you must save and deploy the app.
    • Saving and deploying the app will save your changes and create a new deployment of your application.
    • When including content with curly braces {} in your prompt, such as JSON or other data formats, double each brace ({{ and }}) to prevent it from being interpreted as a variable. This ensures that the content is treated as literal text and maintains its intended structure

Application Types:

AI Assistant Retrieval:

A chain-based application that allows users to connect to a single data source and quickly receive responses specific to the context.

Advanced Agent:

An agent-based application that enables users to connect to multiple tools and perform complex tasks. Users can ask questions related to a specific tool or general inquiries

Basic LLM Assistant

A lightweight application that provides direct interaction with the LLM without additional tools or contextual data. This type serves as a simple mediator between the user and the model, optimizing response time when no extra context is required.


Application status

Another aspect to be aware of is the ability to view your application's status. There are several possible statuses available:

  • Available: The service status will show as "Available" when it's up and ready for use. However, this status can still be "Available" even if there are issues within the application, such as with a model or tool.
  • Pending: This status applies when the application is being deployed. It occurs after the user clicks the "Save & Publish" button but before the application is published.
  • Unavailable: when the service is down.

Ready to start? Follow these straightforward steps to build your application step by step. You can always refer back to this guide as needed for a seamless integration experience within any existing application.

Build your application

  1. Create new app - How to create a new application.
    1. Create a new AI Assistant Retrieval app
    2. Create a new Advanced agent app
    3. Create Basic LLM Assistant Application
  2. Embed your app- How to embed your application anywhere.

Advanced capabilities:

You might find helpful:

Updated about 1 month ago