In the "Deploy your Use Cases" section, you can manage and oversee the deployment of all your AI-driven use cases across Sprinklr. This centralized hub allows you to configure, test, and optimize the application of AI to your specific use cases.

​

Permission Governance

You can access the modules within each product suite depending on the specific permissions granted for each module.

To Configure Use-Case Settings

1. Click the New Tab icon. Under Platform Modules, click All Settings within Listen.

2. Within Manage Customer, search and select AI+ Studio.

   ​

   

3. Click the Deploy your Use-Cases card on the landing page.

   ​

   

4. On the Launchpad window, you will see all AI-driven use cases available to you based on your current enablement and permissions. If you are unable to access a use case despite having the necessary enablement and permissions, please contact the support team for assistance.

   ​

   

5. To configure deployments, select the desired use case. If you've chosen a Sprinklr Provided LLM, you'll notice a Sprinklr Default Deployment already in place, which is view-only and cannot be edited.

   ​

   

   

6. To create a custom deployment, click + Deployment in the top right corner and follow these steps:

   • Name: Choose a meaningful name for your deployment.

   • Priority: Set a priority level for your deployment. Only one deployment can affect a particular record, so the highest-priority deployment will be applied if multiple deployments are applicable.

   • Description: Provide a description that explains the persona and purpose of this deployment.

   • Provider: Select the provider you want to use for this deployment.

   • How to Start: You can either create a deployment from scratch or choose from Sprinklr-provided templates.

   • Filter: Apply filters to target specific records that you want the deployment to influence.

   ​

   

7. Click Next to proceed to the prompt editor, where you can configure the following fields:

   • Base Prompt: Enter clear instructions for the AI model to generate the desired output. For example, you might use a prompt like "Create a numbered list of significant occurrences in the conversation."

   • Hyperparameters: Adjust these settings to fine-tune the model’s behavior and performance according to your specific needs.

   • Placeholders: Define placeholders within your prompts to dynamically insert data during the deployment process. You can add placeholders in the prompt using the “@” symbol.

   ​

   

Testing a Deployment

Once you've entered the prompt and all the necessary placeholders, you can test your deployment with the following steps:

1. Click Test. A modal window will appear for selecting placeholder values.

   

2. Input the values you want to use for testing the deployment, then click Test at the bottom of the modal.

On the right-hand side, the testing pane will display both the request and response. The request shows the exact input sent to the model, and the response displays the model's output based on the provided prompt, hyperparameters, and placeholder values.

If the results meet your expectations, you can deploy the prompt by clicking the "Deploy" button. Alternatively, you can save your work as a draft to continue refining it later.

If you encounter any issues or need additional help, you can consult the Help Center or reach out to the support team for troubleshooting, optimization advice, or any questions you may have.