Hebbrix
Profiles

User Profiles

Build rich user profiles through dialectic conversations. Hebbrix learns user preferences, interests, and personality traits to provide personalized AI experiences.

How Profiles Work

Automatic Learning

Profiles are built automatically from conversations. Every interaction adds to the user's profile.

Dialectic Chat

Use the dialectic endpoint to have focused conversations that extract specific profile information.

Profile Fields

Profiles can store any structured data. Common fields include:

name
preferences
interests
expertise
communication_style
goals
timezone
language
industry
role
personality_traits
custom_fields

Endpoints

Code Examples

Get User Profile

Python
from hebbrix import Hebbrix

client = Hebbrix()

# Get current user's profile
profile = client.profile.get()

print(f"Name: {profile.name}")
print(f"Interests: {', '.join(profile.interests)}")
print(f"Preferences: {profile.preferences}")
print(f"Communication style: {profile.communication_style}")

Dialectic Conversation

Python (Dialectic)
# Start a dialectic chat to learn about the user
response = client.profile.dialectic(
    message="I'm a software engineer interested in AI",
    topic="interests"
)

print(response.assistant_message)
# "That's great! What aspects of AI interest you most -
#  machine learning, NLP, or something else?"

# Continue the conversation
response = client.profile.dialectic(
    message="I'm particularly interested in LLMs and memory systems",
    session_id=response.session_id
)

# Profile is automatically updated with learned interests

Update Profile Manually

Python (Manual Update)
# Set or update profile fields
client.profile.update(
    name="John Doe",
    preferences={
        "theme": "dark",
        "notifications": True,
        "language": "en"
    },
    custom_fields={
        "company": "Acme Inc",
        "role": "Senior Engineer"
    }
)

cURL Example

GET/v1/profile
curl -X GET "https://api.hebbrix.com/v1/profile" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json"

Profile Injection in Chat

When using the chat completions endpoint, profile data is automatically injected:

Python
response = client.chat.completions.create(
    model="gpt-4",
    messages=[{"role": "user", "content": "Recommend a book for me"}],
    features={
        "memory": True,
        "user_profile": True  # Injects profile context
    }
)

# AI knows user's interests and preferences for personalized recommendations

Assistant

Ask me anything about Hebbrix