New capabilities from HAQM Q Business enable ISVs to enhance generative AI experiences

Since its launch, companies have been using HAQM Q Business to improve their employees’ productivity with a generative AI–powered assistant that helps them make better decisions based on company data and information. Employees also use various software applications provided by independent software vendors (ISVs) to complete their tasks. Many ISVs are creating their own generative AI features intended to make their users more productive, but ISVs are often limited to data within their own application, resulting in end users still shifting between applications to complete tasks.

Today, we’re excited to announce new HAQM Q Business capabilities for ISVs. ISVs can now integrate with the HAQM Q index to retrieve data from multiple sources through a single API and customize the design of their HAQM Q embedded assistant.

These new capabilities enable ISVs and application developers to rapidly deploy personalized, AI-powered experiences within their applications, leveraging both enterprise knowledge and user context across multiple software-as-a-service (SaaS) applications, while accelerating their generative AI roadmap with HAQM Q Business capabilities.

Enhance your generative AI features with additional data using the HAQM Q index
With this new capability, ISVs can access content and context from outside their application, helping them to build richer experiences, improve engagement and retention, while complementing their existing generative AI and Retrieval Augmented Generation (RAG) workflows using their preferred large language models (LLMs). Importantly, customers maintain full ownership of their index and have complete control over which applications can access their data.

Software providers register their applications with HAQM Q Business to allow their customers to grant access to their indexed data. After verification, software providers can use this additional data to enhance their built-in generative AI features, delivering more personalized responses to customers. Visit the HAQM Q index for software providers web page to learn more.

After ISVs complete their integration with the HAQM Q index, they have two paths to onboard their customers to use this new, cross-application experience.

1. Onboarding through the ISV’s application — Customers initiate the process through the ISV’s platform. The ISV creates an HAQM Q Business application and index on behalf of each customer. Customers then provide the ISV with credentials to connect additional data sources. In this scenario, the ISV maintains complete control over the onboarding experience and user interface.
2. Onboarding through AWS Management Console – Customers create their HAQM Q Business application directly through the AWS console, where they can connect data sources and grant ISV access to their index. Verified ISVs will be listed as “data accessors” on the HAQM Q Business console. This verification status is granted when the ISV has completed the necessary verification process mentioned above and is ready to launch their customer experience.

Next, we’ll outline the process for a customer to grant a verified ISV access to their existing index.

After customers create their application and add their index, they can grant access to verified ISVs. They can do this by selecting Data accessors in the left navigation panel and then choosing Add data accessor.

On the Add data accessor page, customer will find the list of all verified ISV applications.

After selecting the ISV application, the customer configures what data the ISV can access. The customer also chooses which users will be granted access to the ISV’s updated features.

After granting access, customers must complete the setup by linking their HAQM Q Business application in the ISV’s admin console. Once completed, ISVs can begin retrieving data from the designated index using the SearchRelevantContent API to retrieve data from the index to enrich their generative AI capabilities. Here’s a sample code snippet to use this API:

import boto3
import pprint
qbiz = boto3.client("qbusiness", region_name="us-east-1", **credentials)
 
Q_BIZ_APP_ID = ${Q_BIZ_APP_ID}
 
Q_RETRIEVER_ID = ${Q_RETRIEVER_ID}
 
Q_DATA_SOURCE_ID = ${Q_DATA_SOURCE_ID}
search_params = {
    'applicationId': Q_BIZ_APP_ID,
    'contentSource': {
        'retriever': {
            'retrieverId': Q_RETRIEVER_ID
        }
    },
    'queryText': 'Order coffee API',
    'maxResults': 5,
    'attributeFilter': {
        'documentAttributeFilter': {
            'andAllFilters': [{
                'equalsTo': {
                    'name': '_data_source_id',
                    'value': {
                        'stringValue': DATA_SOURCE_ID
                    }
                }
            }]
        }
    }
}
search_response = qbiz.search_relevant_content(**search_params)

Customize the design of the embedded assistant
HAQM Q embedded is a capability that helps ISVs extend HAQM Q Business to their end users by embedding an AI-powered assistant into their user interface. This capability helps ISV users complete various tasks, such as summarizing documents and answering questions.

Now, software providers have the option to customize the embeddable generative-AI assistant user interface (UI) with HAQM Q embedded to match their corporate branding. To get started, select HAQM Q embedded in the left navigation panel and choose Customize web experience.

On this page, select Theme to start customizing generative AI assistant UI look and feel, such as configuring the assistant name, welcome message, color scheme, and logo.

Available today
The HAQM Q index and HAQM Q embedded with customizable UI are generally available today in the US East (N. Virginia) and US West (Oregon) AWS Regions, with availability in additional AWS Regions coming soon.

ISVs can now use HAQM Q Business features to innovate and enhance their user experiences with powerful AI capabilities. To learn more about how ISVs can enhance their applications, visit HAQM Q Business page for software providers.

Happy coding!

— Donnie