Overall Workflow

Understanding how the Ontbo API integrates with your AI system to provide personalized user context

The Ontbo API is designed to be seamlessly integrated into AI systems that handle conversational data with users. It processes conversations as scenes (individual conversation sessions) and organizes user information into profiles (comprehensive user knowledge databases).

System Architecture

Your AI System

Chatbot, Assistant, or AI Application

Ontbo API

Context & Memory Management

Data Flow: Your AI system sends conversational data to Ontbo API, queries user context, and receives personalized information to enhance responses.

1

Profile Creation & Data Ingestion

Profiles are automatically created when you send conversational data to the Ontbo API. The system intelligently extracts relevant user information and stores it in structured profiles.

When to Send Data

    You can choose send data either:

  • • At the end of each conversation session
  • • During regular data synchronization
2

Profile Querying & Context Retrieval

Query user profiles to retrieve relevant context for personalized responses. The API offers two powerful querying approaches:

Direct Questions

Ask specific questions about the user using the facts query endpoint. The Ontbo API will explore the profile database to find the most accurate answer to your question.

GET /profiles/{profile_id}/facts/query
Query parameter: "What is the user's job?"
Example Response:
{ "result": "The user works as a software engineer at a tech startup." }
Behavior: The API searches through all stored user information to find facts that directly answer your specific question.

Context Analysis

Forward user requests for intelligent context matching. The Ontbo API will analyze the user input and autonomously find the most relevant information to answer the question.

GET /profiles/{profile_id}/context
Input parameter: "I am planning a trip to Paris, what places should I visit?"
Example Response:
{ "context": "The user enjoys cycling and outdoor activities. They mentioned loving nature spots and prefer less crowded areas. Based on their cycling interest, they might enjoy bike-friendly routes and parks in Paris." }
Behavior: The API analyzes the user's request and intelligently extracts relevant context from their profile to provide personalized recommendations.

How to Use the Results

The retrieved context from either method should be embedded in your AI system's prompt. This enables your AI to provide more personalized, relevant, and contextually aware responses to users, making each interaction feel tailored to the individual user's preferences and history.

3

Enhanced AI Responses

Embed the retrieved context in your AI system's prompt, enabling more personalized, relevant, and contextually aware responses to users.

Key Benefits

Personalized Responses: AI responses tailored to each user's preferences and history

Context Awareness: Maintains conversation context across sessions

Scalable Architecture: Handles multiple users and conversations efficiently

Easy Integration: Simple REST API that works with any AI system