GibsonAI Database Integration

Learn how to use Memori with a serverless database in GibsonAI platform for persistent memory storage.

Overview

GibsonAI provides a serverless MySQL/PostgreSQL compatible database platform that seamlessly integrates with Memori for persistent memory storage. This integration allows you to maintain conversation memory across sessions with zero database management overhead.

Quick Setup

1. Create a GibsonAI Account

1. Visit https://app.gibsonai.com/
2. Sign up for a FREE account

2. Create a Database Project

1. Click "Create New Project" in the GibsonAI dashboard
2. Use a prompt like: "Create an empty database"

3. Get Your Connection String

1. Navigate to the Databases tab in your GibsonAI project
2. Choose your environment:
   • Development: For testing and development
   • Production: For live applications
3. Copy the MySQL connection string provided

The connection string format looks like:

mysql+mysqlconnector://username:password@mysql-assembly.gibsonai.com/database_name

4. Install Dependencies

pip install memorisdk openai mysql-connector-python

5. Set Environment Variables

export OPENAI_API_KEY="your-openai-api-key-here"

Basic Usage

Simple Integration

from openai import OpenAI
from memori import Memori

# Initialize OpenAI client
openai_client = OpenAI()

# Initialize Memori with GibsonAI database
memori = Memori(
    database_connect="mysql+mysqlconnector://your_username:your_password@mysql-assembly.gibsonai.com/your_database",
    conscious_ingest=True,
    auto_ingest=True,
)

# Enable memory tracking
memori.enable()

# Use with any LLM conversation
response = openai_client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Hello, I'm learning Python!"}]
)

print(response.choices[0].message.content)