Push JSON into RushDB. Get semantic recall, graph traversal, analytical queries, and live schema over the same memory.
Typical stack
Memory split across systems.
App records, session state, embeddings, metadata, tool output, and relationships drift across separate stores.
Application DB
facts, history, metadata
Redis
session state, cache
Vector DB
embeddings, similarity search
Your code owns:
sync · embed · index · join · retry
With RushDB
Push JSON. Get connected memory.
RushDB keeps facts, events, tool output, embeddings, and relationships in one queryable memory layer.
Agents can recall by meaning, traverse relationships, and aggregate over the same memory.
RushDB keeps typed fields, searchable text, vector embeddings, and relationships together as data arrives. Agents can recall by meaning, traverse relationships, and aggregate over memory without a sync pipeline.
Models, frameworks, MCP clients, and workflows will change. RushDB keeps the records, relationships, semantic indexes, and live schema outside any one session.
Change the stack without rebuilding memory.
Memory lives outside any one model, framework, MCP client, or provider session, so the next agent can continue from the same durable context.
Durable memory layer
Every provider reads the same records, relationships, and searchable context.
SDK, MCP, and application workflows share one durable memory layer.
Keep context available without replaying transcripts or migrating session state.
Install
Add the SDK for TypeScript, Python, or use the REST API directly — no extra dependencies.
Push
Write any JSON object. RushDB parses field types, builds the graph structure, and indexes vectors server-side on the same write.
Recall
Query by meaning, by graph relationship, or both in a single call using the same SearchQuery shape every time.
RushDB is domain-agnostic: the same graph memory layer adapts to agents, RAG, apps, sales, support, and healthcare without hard-coding a new store for each workflow.
Decisions, tool outputs, and entities survive across runs as reusable records.
Recall prior work by meaning, scoped to a session, agent, or task.
Multiple agents read and write the same durable memory graph.
RushDB turns messages, tool outputs, documents, entities, and events into queryable context without a separate graph pipeline, vector sync job, or agent-specific storage layer.
Field notes
RushDB is built for the memory requirements production AI teams actually face: connected context, live schemas, semantic retrieval, and a storage model agents can safely inspect.
“In a few minutes, RushDB can go from Docker Compose to Python SDK to nested JSON turning into a real graph. For developer education, that matters: builders can see the data land in the visualizer and Neo4j Aura, not just read about it.”
Fractional DevRel | Video-First Developer Content & Community Growth | ex-Neo4j
NODES 2025 · hosted by Neo4j
A Neo4j-hosted conference session by Artemiy Vereshchinskiy, Founder, RushDB, showing RushDB in action for AI-native apps, self-aware graphs, and practical graph-backed agent workflows.
Free includes 100K KU/month. Pro starts at $30/month. Reads are free; writes are usage-based. Full pricing details →
Free
Evaluate RushDB in production conditions
Get started instantly:
~3,000 records with 10 fields each
No time limits, no feature restrictions
Connect to your own Neo4j or Aura
Pro
Perfect for production applications
Everything in Free, plus:
100× Free plan — ~300,000 records with 10 fields each
Continue beyond 10M KU at $3 / M KU
then $10 / seat
Scale
For high-volume and data-intensive apps
Everything in Pro, plus:
Usage-based — pay only for what you consume