🎛️APP

Overview of ZBrain apps

ZBrain enables users to create LLM-powered applications tailored to their specific needs. These applications, such as customer support systems, chatbots or question-answering tools, can be personalized in terms of features, responses, and operations, making them highly adaptable to various use cases. Once developed, these apps can be seamlessly integrated into an organization's workflows.

Types of apps

ZBrain offers flexibility in the types of applications that can be built, catering to different user requirements:

Public apps

Public apps are accessible to anyone with the link. These apps are designed for a broader audience, making them suitable for customer-oriented use cases like service chatbots.

Private apps

Access to private apps is restricted and is controlled by the app admin. These apps are suitable for internal or restricted use cases involving sensitive information or specific functionalities.

Orchestration methods

Users can choose from different orchestration methods to define how their applications will handle and respond to queries:

Knowledge base (Basic)

This method uses a predefined knowledge base to answer queries, leveraging stored information for straightforward, information-driven responses.

Flow (Advanced)

Flow allows for the design of custom interactions and processes, enabling complex query handling. This method is ideal for advanced users who need to implement specific logic and sequences.

Customization options

ZBrain apps offer extensive customization options to ensure they meet specific business needs and enhance user interaction:

Knowledge base integration

Users can connect one or multiple knowledge bases to their app, facilitating detailed data queries and ensuring comprehensive information retrieval.

Schema selection

Linked schemas can be selected to structure the knowledge base, optimizing the retrieval and presentation of information.

Instructions and guidelines

Users can define specific instructions for the app to follow during interactions. These guidelines can include directives to provide detailed answers, use natural language, and tailor responses to the conversation's context.

Manual configuration

Users can manually modify the application's settings to customize default configurations by adjusting parameters such as model, temperature, score threshold, context length, etc. Once customized, these configurations will be saved and applied to the application in subsequent uses.

Note: For more detailed information on each parameter, please refer to How to create a new app? | ZBrain Documentation

Appearance

Users can customize the visual and interactive aspects of the app by configuring the welcome message, app name and description, sample questions, logo, theme, bot name, and icons. These settings enhance the user experience and align the app with your branding.

Capabilities of ZBrain apps

Integration and sharing

Once the app is created, users can share it via a link or integrate it into their website using a conversational widget. This allows easy access to the app's functionalities and extends its reach.

Collecting user information

The app can prompt users to provide information during interactions. This feature enabes specifying the type of information to be collected, enhancing both user engagement and data collection.

Auto configuration

Users can automatically configure their apps, allowing the system to select the best settings. This ensures optimal performance tailored to their app's needs.

Testing and performance

Users can test their app using sample questions to evaluate its accuracy and performance. Detailed reports provide insights into the app's effectiveness, including accuracy scores, response times, and user satisfaction. This ensures the app meets desired performance standards and provides valuable feedback for further optimization.

Query history and reports

The query history and reports sections offer valuable insights into user interactions and app performance. Users can track session details, monitor usage patterns, and analyze performance metrics. This data helps make informed decisions to optimize the app and enhance user experience.

Exporting reports

Users can easily export reports by selecting the desired data fields and generating a CSV file. The file provides a convenient way to document, analyze and share data.