Connect any system. Its structure becomes a draft ontology, and its data is queried in place.

Request a demo

Swipe up to continue

Connectors for the systems you run. PostgreSQL, MongoDB, ClickHouse, Snowflake, Databricks, HubSpot, Stripe, Notion and Confluence, each ready to connect.

Request a demo

Swipe up to continue

Connect, and its structure becomes a draft ontology. Only the shape is read: the catalog or the spec. Rows are never read and never stored.

Request a demo

Swipe up to continue

One vocabulary for every source. SQL, REST and GraphQL types land on the same six datatypes: string, integer, decimal, boolean, date and datetime.

Request a demo

Swipe up to continue

Names you can read. Classes come out singular and PascalCase, relationships as UPPER_SNAKE.

Request a demo

Swipe up to continue

Ask in ontology, run as native queries. A request shaped like your ontology becomes the query each system speaks, from SQL to GraphQL.

Request a demo

Swipe up to continue

Answers come back unified. Whatever each source returns is converted back into the same shape, ready to use together.

Request a demo

Swipe up to continue

Read-only, and audited. Read-only access, SSO, role-based access and a full audit trail. Query in place, no copies.

Request a demo

Swipe up to continue

More about Connector. Structure mapping today and what it takes to run it.

Request a demo

Swipe up to continue

Structure mapping today

Structure mapping today. Each source has its own mapper, and all of them land in the same model.

SourceWhat is readWhat it becomes
PostgreSQL postgresql://The catalog only: information_schema and pg_catalog, for tables, columns, keys and comments. The public schema, base tables.Tables become classes, columns become typed properties, keys become relationships. Comments become definitions.
OpenAPI openapi.json 3.xOne request for the spec, in JSON.Schemas with properties become classes.
GraphQL introspectionOne introspection query.Object types become classes; their fields become data or object properties.
Built for production

Built for production

Fetched when it is asked for

Data is collected from the source at request time.

Rebuild without duplicates

Re-import in place and the ID stays the same.

Read it back as clean JSON

Every domain and range joins straight to the class list.

GET /instances/hr-7f3a/schema
{
  "classes": [
    { "label": "Employee",
      "data_properties": [
        { "name": "salary", "datatype": "decimal" },
        { "name": "location", "datatype": "string" }
      ] },
    { "label": "Department", … }
  ],
  "object_properties": [
    { "label": "HAS_DEPARTMENT",
      "domain": "Employee", "range": "Department" }
  ]
}

A safe lifecycle

Soft delete, restore, then delete for good.

Errors that say what happened

A distinct status for every failure.

Live API docs

Try every call from the browser.