Examples

from hippocamp import MemoryEngine
import os

# Initialize the memory engine
engine = MemoryEngine(
    api_key=os.getenv("OPENAI_API_KEY"),
    qdrant_host="localhost",
    qdrant_port=6333
)

print("Memory engine initialized!")

# Create a session for the user
user_id = "user_123"
session_id = "chat_session_1"

session = engine.create_session(
    user_id=user_id,
    session_id=session_id
)

print(f"Session created: {session_id}")

# User sends a message
user_message = "Hi! I prefer dark mode and I'm interested in AI."

# Store the user message
session.add_message(
    role="user",
    content=user_message
)

# Store user preferences as a separate memory
engine.store(
    content="User prefers dark mode interface",
    user_id=user_id,
    metadata={"type": "preference", "category": "ui"},
    importance=0.8
)

engine.store(
    content="User is interested in AI and machine learning",
    user_id=user_id,
    metadata={"type": "interest", "category": "topics"},
    tags=["ai", "ml", "technology"]
)

print("Messages and preferences stored!")

# When generating a response, get relevant context
query = "What does the user like?"

# Search for relevant memories
memories = engine.search(
    query=query,
    user_id=user_id,
    limit=5
)

print("Relevant memories found:")
for memory in memories:
    print(f"  - {memory.content} (relevance: {memory.relevance:.2f})")

# Get conversation context
context = session.get_context(limit=10)
print(f"\nConversation context: {len(context)} messages")

from hippocamp import MemoryEngine
from openai import OpenAI

class MemoryEnabledChatbot:
    def __init__(self, api_key: str):
        self.engine = MemoryEngine(api_key=api_key)
        self.openai = OpenAI(api_key=api_key)
        self.user_id = None
        self.session = None
    
    def start_session(self, user_id: str, session_id: str):
        self.user_id = user_id
        self.session = self.engine.create_session(user_id, session_id)
    
    def chat(self, user_message: str) -> str:
        # Store user message
        self.session.add_message(role="user", content=user_message)
        
        # Get relevant memories
        memories = self.engine.search(
            query=user_message,
            user_id=self.user_id,
            limit=5
        )
        
        # Build context
        memory_context = "\n".join([m.content for m in memories])
        
        # Generate response with context
        response = self.openai.chat.completions.create(
            model="gpt-4",
            messages=[
                {"role": "system", "content": f"User context:\n{memory_context}"},
                {"role": "user", "content": user_message}
            ]
        )
        
        assistant_message = response.choices[0].message.content
        
        # Store assistant response
        self.session.add_message(role="assistant", content=assistant_message)
        
        return assistant_message

# Usage
bot = MemoryEnabledChatbot(api_key="your_key")
bot.start_session("user_123", "chat_1")
response = bot.chat("What's the weather like?")
print(response)