Skip to main content

Hello LLM

Initiation​

CLI Command: para init --template hello_llm

Why this project?​

This project demonstrates LLM integration with Paranet with sample approaches for building production LLM applications.

What this kit showcases​

  • Direct LLM integration using Azure OpenAI via Python actors.
  • Three interaction patterns:
    • Raw conversational interface
    • Medical assistant with specialized prompts
    • Template-based email classification and routing
  • Actor-based LLM orchestration using Paraflow skills.
  • Environment-based configuration for secure API management.

How to run it​

1 Set up environment variables​

Create a .env file with Azure OpenAI credentials:

AZURE_OPENAI_API_KEY=<your key here>
AZURE_OPENAI_ENDPOINT=<your endpoint here>

2 Deploy Paranet services​

para devkit login
para docker deploy node
para docker deploy package

3 Test interaction patterns​

  1. Navigate to Paracord UI → Actor Hub.
  2. Try each pattern:

Direct LLM Conversation (chatgpt/direct_chat):

  • Enter any question in the prompt field
  • Get raw conversational responses without system context

Medical Assistant (chatgpt/medical_chat):

  • Enter healthcare-related questions
  • Receive professional medical assistant responses with specialized context

Email Classification (email_classifier/prioritize):

  • Fill in email details (subject and body)
  • LLM categorizes the email and Paraflow routes based on classification