Creating an Assistant

Navigate to Assistants

Begin by navigating to the Assistants section in the main navigation menu.

Initiate Assistant Creation

Click on the “Add new assistant” button to start the process of creating a new assistant.

Select your LLM

Select the right LLM for your assistant:

• Choose the appropriate model from the model provider list. The platform supports various model providers including:
  • OpenAI
  • Azure OpenAI
  • Anthropic
  • Google AI
• For each provider, you’ll need to select a specific model (like GPT-4o for Azure OpenAI as shown in the image).
• Make sure you’ve already configured credentials for your chosen provider in the Integrations section.

Provide Assistant Instructions

Define the detailed instructions that guide the assistant’s behavior. This will be the System prompt used by the assistant.

Add Function Tools

Add function tools to extend your assistant’s capabilities. The platform supports several tool types:

• Knowledge Retrieval: Access and search documents from your knowledge base on Rapida
• API Request: Make external API calls to fetch or push data in realtime
• Endpoint (LLM Call): Trigger other LLM services for your special use cases
• Put On Hold: Pause conversation flow if you want to implement call Hold
• End Of Conversation: Terminate the call when conversation is over.

Finalise Name

Enter the basic information for your assistant:

• Name: Give your assistant a descriptive name
• Description: Provide a brief explanation of the assistant’s purpose

Click “Create Assistant” when done.You’ll see a confirmation that your assistant has been created successfully. You can now start using your new assistant or make additional configurations. To configure voice on your Twilio Phone for this assistant, visit please refer to the Twilio Phone Number Deployment Manual.