Skip to main content

Quickstart

Go from zero to working memory in under 10 minutes. Install the SDK, create an instance, and ingest your first memory.

Core Concepts

Understand how Synap structures memory — scopes, entities, memory types, and the ingestion-to-retrieval lifecycle.

SDK Reference

Complete reference for the Python SDK, including initialization, ingestion, context fetch, and configuration.

Dashboard Guide

Manage instances, configure memory architecture, monitor ingestion pipelines, and collaborate with your team.

​
What you can build

Synap handles the hard parts of agent memory so you can focus on building great AI experiences:
  • Personalized chatbots that remember user preferences, past interactions, and evolving context across sessions
  • Context-aware support agents that recall customer history, previous tickets, and account details without re-asking
  • Agents that learn over time by extracting and retaining facts, preferences, emotions, and temporal events from every conversation
  • Multi-user, memory-scoped applications where each user, customer, or organization has isolated, structured memory boundaries

​
How it works

Synap’s memory pipeline is designed to be invisible to your users and effortless for you as a developer.
1

Ingest conversations and documents

Send raw conversations, documents, or any text to Synap through the SDK. The ingestion pipeline runs asynchronously — your application never blocks.
await sdk.memories.create(
    document="User: I prefer dark mode and concise responses.\nAssistant: Got it!",
    document_type="ai-chat-conversation",
    user_id="user_123"
)
2

Synap extracts structured knowledge

The ingestion pipeline automatically extracts facts, preferences, episodes, emotions, and temporal events from your content. Entities are resolved across conversations so “John”, “John Smith”, and “my manager” all map to the same person.
3

Retrieve relevant context for any conversation

When your agent needs context, query Synap with a search and get back ranked, relevant memories scoped to the right user, customer, or organization.
context = await sdk.conversation.context.fetch(
    conversation_id="conv_456",
    search_query=["user preferences", "past interactions"],
    max_results=10
)
Synap architecture overview showing the SDK, Cloud, and Dashboard components

​
Architecture overview

Synap is composed of three main layers that work together:
ComponentRoleHow you interact
SDKRuns in your application. Handles ingestion, retrieval, authentication, and streaming.pip install maximem-synap
Synap CloudManaged backend. Runs the ingestion pipeline, stores memories, handles entity resolution, and serves retrieval queries.Fully managed — no infrastructure to deploy
DashboardWeb UI for managing instances, configuring memory architecture, monitoring pipelines, and team collaboration.synap.maximem.ai
Synap is async-first. The SDK uses Python’s asyncio for non-blocking operations, and the ingestion pipeline processes memories asynchronously so your application stays responsive.