Overview

Add persistent memory to your AI app in 5 minutes

Databaset gives your AI apps long-term memory. Store user preferences, conversation context, and facts. When you need them again, recall by meaning, not keywords.

Prerequisites

• Node.js 18+ or Python 3.10+
• A Databaset account
• An API key from the dashboard

Step 1: Install SDK

npm install @databaset/sdk

pip install databaset

Step 2: Get your API key

1. Go to Dashboard → API Keys → Create Key
2. Copy your key and set it as an environment variable:

export DATABASET_API_KEY="db_..."

Step 3: Initialize client

import { Memory } from '@databaset/sdk'

const memory = new Memory() // reads DATABASET_API_KEY automatically

from databaset import Memory

memory = Memory()  # reads DATABASET_API_KEY automatically

export DATABASET_API_KEY="db_your_key_here"

Step 4: Store your first memory

await memory.store({
  userId: 'user_123',
  text: 'User prefers dark mode and speaks Hindi'
})

memory.store(
    user_id="user_123",
    text="User prefers dark mode and speaks Hindi"
)

Step 5: Recall memory

const context = await memory.recall({
  userId: 'user_123',
  query: 'what are user preferences?'
})

console.log(context)
// "User prefers dark mode and speaks Hindi"

context = memory.recall(
    user_id="user_123",
    query="what are user preferences?"
)
print(context)

Step 6: Use with Claude or GPT

Inject recalled context into your AI prompt for personalized responses.

const context = await memory.recall({
  userId: 'user_123',
  query: userMessage
})

const response = await anthropic.messages.create({
  model: 'claude-sonnet-4-6',
  messages: [
    {
      role: 'user',
      content: `Context: ${context}\n\nUser: ${userMessage}`
    }
  ]
})

What's next

• API Reference: full REST API documentation
• Build a chatbot with memory: end-to-end example
• Node.js SDK: complete SDK reference
• Self-hosting guide: run Databaset on your infrastructure