NamiDB
Status

All systems
as expected.

A live mirror of the surfaces that exist today and the engine phases behind them. The engine reached its first stable release under semantic versioning, so breaking changes now bump the major version. While the cloud is in closed beta, uptime reflects internal monitoring; measured SLA data publishes the day we go to public beta.

All public surfaces are operational.

Services

  • namidb.com marketing site

    Static. Served from the CDN edge.

    Operational

  • Waitlist intake

    Form submissions accepted. Data stays with us.

    Operational

  • NamiDB CLI + Rust workspace

    v1.1.0. Install via cargo. Workspace-wide tests + clippy + fmt clean.

    Shipped

  • NamiDB Python bindings

    v1.1.0. pip install namidb. abi3 wheels for Linux, macOS arm64, Windows x86_64.

    Shipped

  • NamiDB Server (Rust daemon + Docker)

    v1.1.0. namidb-server binary + container image. REST on /v0/*, plus a Bolt listener for Neo4j drivers.

    Shipped

  • NamiDB MCP server

    v1.1.0. namidb-mcp. Read-only graph access over stdio for agents like Claude Code.

    Shipped

  • Documentation (docs.namidb.com)

    Reference docs + guides land alongside the Cloud public beta.

    Planned

  • NamiDB Cloud (managed multi-tenant)

    Closed beta on namidb.com. Design partners only. Request access from the homepage.

    Closed beta
Engine phases

Where the work
actually is.

The plan ships in phases. Each gate has an explicit definition of done. When we close one, the threshold number is published here in the same breath. When we miss one, the slip is published here too.

  • Phase 0: Architectural validation continuous

    Every non-trivial change starts as an RFC. Two dozen in the open so far.

    In progress

  • Phase 1: Storage engine 100%

    On-disk format stable under semantic versioning as of v1.0. LSM tree + manifest CAS over object storage. Cold p50 < 500 ms, warm p50 < 10 ms, ingest >= 10 K nodes/s.

    Done

  • Phase 2: Query engine v1.0 stable

    Cypher surface stable under semver: cost-based optimizer, subqueries, DDL, 7 graph algorithms, vector and BM25 hybrid search. Four LDBC IC queries benched at scale=1.0, 3-5x over Kuzu on IC02 / IC08. More queries land in the next sweep.

    Done

  • Phase 3: Multi-tenant Cloud ~30%

    Closed beta on namidb.com, design partners only. Public beta gated on docs + SLA targets.

    In progress

  • Phase 4: GraphRAG SDK 0%

    First-class hybrid vector + graph retrieval on top of the Bolt + REST surface.

    Planned

  • Phase 5+: Scale + long-term support 0%

    Wire protocol freeze. Long-term support commitments.

    Planned

Subscribe

Status changes will be mirrored to the waitlist once the cloud is in public beta. Until then, expect quiet on this page. Incidents are handled internally with our design partners.

Incident reports

Every public incident lands as a postmortem in /research within seven days. Including the ones that embarrass us.

Run it on your own bucket.

Early access to the Cloud is open. One launch email when the engine is ready, never spam.

Request early access