Rushdb

RushDB is a graph and vector database designed as a persistent memory layer for AI agents and applications. Developers push arbitrary JSON and automatically get typed, searchable, relationship-aware records without defining a schema or running migrations. Built on Neo4j, it combines graph relationships, vector embeddings, full-text search, and live schema discovery into one queryable memory layer. RushDB targets AI developers building agentic systems, RAG pipelines, and applications that need to keep state, decisions, and tool output available across sessions. It is particularly suited for use cases like customer support memory, sales CRM memory, transaction monitoring, and multi-agent incident response. The workflow is simple: push JSON via REST API or SDKs, and RushDB reconstructs the graph by inferring property types, linking nested objects, and suggesting relationships. The platform offers a REST API, TypeScript and Python SDKs, and a native MCP server. Features include native semantic search, ontology-aware querying, MCP with OAuth, and bring-your-own Neo4j. Pricing is based on knowledge units (KU) written, with reads always free. RushDB also supports SSO authentication and a dashboard with query lab, saved queries, and performance optimization. RushDB differs from traditional databases and vector stores by eliminating the need for separate sync pipelines for embeddings, metadata, and relationships. It supports ACID transactions, open-source deployment, and self-hosted or cloud options. Compared to solutions like Mem0 or Pinecone, RushDB offers a unified graph+vector approach with built-in relationship context.