AIPersona

Quick Start Guide

Get up and running with AIPersona in under 10 minutes. Create your first AI persona and start having conversations.

Lightning Fast

Create and deploy AI personas in minutes, not hours.

Multi-Provider Support

Works with OpenAI, Anthropic, Google, and more LLM providers.

Tip
This guide shows both dashboard usage and HTTP API calls for external integrations.

Prerequisites

Before we begin, make sure you have:

  • An AIPersona account (sign up at the top right)
  • An organization created (done automatically on first login)
  • At least one LLM provider configured (Settings → LLM Configuration)
  • For API access: a way to make HTTP requests (curl, Postman, or your favorite programming language)

1
Get Your API Key

For external integrations, you'll need an API key to authenticate your requests:

  1. Navigate to Settings → API Keys in your dashboard
  2. Click Create New Key
  3. Give your key a descriptive name (e.g., "My First Persona App")
  4. Copy the generated API key and store it securely
  5. Note your Organization ID from the Settings page
Warning
Your API key provides access to your organization's resources. Keep it secure and never commit it to version control.

Required Headers for API Calls:

Bash
X-API-Key: your-api-key-here
X-Organization-ID: your-organization-id
Content-Type: application/json

2
Create Your First Persona

Create an AI persona with a specific personality and purpose. You can do this via the dashboard or API.

Using the Dashboard

  1. Click Personas in the sidebar
  2. Click Create Persona
  3. Fill in the basic info: name, description, and type
  4. Configure your LLM provider and model (e.g., OpenAI GPT-4o)
  5. Write a system prompt or use the Structured Prompt Builder
  6. Optionally enable tools like Web Search or PDF Processor
  7. Click Create Persona
Note
The system prompt defines how your AI persona behaves. Be specific about the role, tone, and constraints you want. Use the Structured Prompt Builder for more organized prompts.

Example Response:

JSON
{
  "result": {
    "data": {
      "id": "clx1abc123xyz",
      "name": "Customer Support Agent",
      "description": "Handles customer inquiries professionally",
      "type": "CUSTOMER_SERVICE",
      "model": "gpt-4o",
      "llmProvider": "OPENAI",
      "createdAt": "2025-01-15T10:30:00.000Z",
      "organizationId": "cmdhtumx9000nus412v5qezme"
    }
  }
}

3
Start a Conversation

Create a chat session with your persona. Each session maintains its own conversation history.

Using the Dashboard

  1. Click Chat in the sidebar
  2. Click New Chat
  3. Select your persona from the dropdown
  4. Optionally give the session a title
  5. Start typing your message!

Example Response:

JSON
{
  "result": {
    "data": {
      "id": "session_xyz789abc",
      "title": "Support Inquiry",
      "personaId": "clx1abc123xyz",
      "platform": "API",
      "status": "ACTIVE",
      "createdAt": "2025-01-15T10:35:00.000Z"
    }
  }
}

4
Send Messages

Send messages to your chat session and receive AI-generated responses.

Bash
curl -X POST "https://aipersona.dsethiopia.org/api/trpc/chat.sendMessage" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: your-api-key" \
  -H "X-Organization-ID: your-org-id" \
  -d '{
    "json": {
      "chatSessionId": "session_xyz789abc",
      "content": "Hello! I need help with my account login.",
      "role": "USER"
    }
  }'

Example Response:

JSON
{
  "result": {
    "data": {
      "id": "msg_abc123",
      "chatSessionId": "session_xyz789abc",
      "role": "ASSISTANT",
      "content": "Hello! I'd be happy to help you with your account login. Could you please tell me what specific issue you're experiencing? For example:\n\n1. Are you getting an error message?\n2. Did you forget your password?\n3. Is your account locked?\n\nOnce I understand the issue better, I can guide you through the solution.",
      "createdAt": "2025-01-15T10:36:00.000Z"
    }
  }
}

What's Next?

Enhance Your Persona

  • • Upload documents to the knowledge base
  • • Enable tools (Web Search, Calculator, PDF Processor)
  • • Use Structured Prompts for better control
  • • Configure embedding models for RAG

Explore the API

Tip
Enable Agent Mode on your personas to allow them to use tools autonomously for complex tasks like web searches and calculations.