Hindsight - Agent Memory System for AI

Hindsight is an agent memory system built for AI agents that need to learn, not just retrieve past chat snippets. Instead of treating memory as plain vector search, it organizes facts, observations, and mental models so agents can answer temporal and multi-hop questions with higher accuracy.

Key Features

• Retain / Recall / Reflect: Core operations for storing knowledge, searching memory, and reasoning over consolidated observations.
• Memory Hierarchy: World facts, experience facts, observations, and mental models keep long-term knowledge structured.
• Benchmark-Proven Accuracy: Reports state-of-the-art LongMemEval performance for long-term conversational memory.
• Simple Integrations: Drop-in LLM wrapper plus SDKs for Python, TypeScript, Go, and CLI workflows.
• Self-Host Friendly: Docker image, compose/Kubernetes paths, and support for external PostgreSQL or Oracle storage.
• Wide Ecosystem Fit: Integrations across LangGraph/LangChain, CrewAI, AutoGen, Claude Code, Cursor, n8n, MCP, and more.

Why Choose Hindsight?

• Your agents forget context between sessions and need durable memory.
• Vector RAG alone cannot handle temporal or multi-hop reasoning well.
• You want an open, self-hostable memory layer with production clients.
• You need one memory bank model that works across multiple agent frameworks.

Docker Deployment

Quick start with Docker:

export OPENAI_API_KEY=sk-xxx
docker run -it --pull always --name hindsight --restart unless-stopped \
  -p 8888:8888 -p 9999:9999 \
  -e HINDSIGHT_API_LLM_API_KEY=$OPENAI_API_KEY \
  -v hindsight-data:/home/hindsight/.pg0 \
  ghcr.io/vectorize-io/hindsight:latest

API defaults to http://localhost:8888 and UI to http://localhost:9999.

Getting Started

1. Run the Docker image or compose stack.
2. Configure your LLM provider (openai, anthropic, gemini, ollama, and others).
3. Install a client (pip install hindsight-client or npm install @vectorize-io/hindsight-client).
4. Call retain/recall/reflect from your agent, or wrap your LLM client for automatic memory.

Full Setup Guide

For the complete Docker deployment walkthrough, read: Hindsight Docker Deploy Guide.