offdataai
Data catalog

A data catalog that knows what your tables mean

Every system, table and column you connect, which business concepts use them and where the same data lives twice. Built from metadata only, searchable by your team and by AI agents.

Catalog your data Book a demo
  • From $1/month
  • 8 SQL targets
  • 174 templates
Connected · PostgreSQLMetadata only
PostgreSQLSnowflakeDatabricksrows read: 0
sales.orders14 cols
sales.order_items9 cols
crm.accounts22 cols
catalog.products17 cols
ref.currency4 cols

Suggested relationships

order_items.order_id
→ orders.order_id
98% confidentreview
orders.cust_ref
→ accounts.account_id
91% confidentreview
Mirror 1:1Remodel as star
Reads tables, columns, keys and comments, never rows, and suggests the joins nobody declared.
Watch the demo

See it in action

A narrated walkthrough, recorded in the real app.

How it works

Three steps, minutes not months

1

Describe or import

Answer a short interview, open a template, draw on the canvas or connect a live database.

2

Review a validated model

Keys, grain, history and PII are checked before any code is written. Refine it in plain English.

3

Ship and keep it current

Native DDL, dbt, seeds, docs and an ontology. Push to Git, diff versions, generate migrations.

Overview

Why teams choose offdataai

app.offdataai.com
Connected systems in the catalog, with tables classified and keys suggested

Connected systems in the catalog, with tables classified and keys suggested

  1. 01

    Most data catalogs answer one question well: what tables do we have?

  2. 02

    The harder questions come next.

  3. 03

    Which of the three customer tables is the real one?

  4. 04

    Which dashboards and metrics depend on this column before I rename it?

  5. 05

    Is this field personal data?

  6. 06

    And now, for AI agents: which table should I query to answer this question at all?

  7. 07

    offdataai's data catalog is built for those follow-up questions because it is connected to the rest of your data model.

  8. 08

    It gathers every system, table and column from your connections: PostgreSQL, MySQL, Snowflake, BigQuery, Amazon Redshift, Databricks Unity Catalog, including foreign catalogs federated through Lakehouse Federation, and HubSpot.

  9. 09

    It then indexes which ontology concepts are bound to each table and column.

  10. 10

    "Who uses this column?" becomes a lookup instead of a hunt through code.

  11. 11

    It also spots duplication.

  12. 12

    When a Postgres table is mirrored into a lakehouse, the catalog links the copy to its original, so you can see where the same data lives twice and the semantic layer can choose the cheapest copy to query.

  13. 13

    All of this comes from metadata: names, types, keys and comments.

  14. 14

    The catalog never reads a row of your data.

  15. 15

    The catalog is one part of a connected stack rather than a separate product to roll out.

  16. 16

    The ontology gives tables business meaning, the semantic layer turns questions into SQL, and the MCP server lets agents search the catalog with search_catalog before they query anything.

  17. 17

    For data modelling teams, that means the inventory, the meaning and the governed access live in one place.

Real output

The code it writes, not a sketch

This is the shape of what offdataai generates here, straight from the validated model. Copy it, download it, or push it to Git.

  • Every connected system
  • Who uses this column?
  • Copies linked to originals
mcp · search_catalogjson
1// search_catalog("customer email")
2{
3  "results": [
4    {
5      "system":  "Snowflake · acme_prod",
6      "table":   "ANALYTICS.SALES.DIM_CUSTOMER",
7      "column":  "EMAIL",
8      "type":    "VARCHAR",
9      "classification": "PII · masked (hash)",
10      "used_by": ["Customer.email  (ontology: acme-sales)"]
11    },
12    {
13      "system":  "Postgres · crm",
14      "table":   "public.contacts",
15      "column":  "email_address",
16      "copy_of": null,
17      "used_by": ["Customer.email  (ontology: acme-sales)"]
18    },
19    {
20      "system":  "Databricks · lakehouse",
21      "table":   "crm_mirror.public.contacts",
22      "copy_of": "Postgres · crm · public.contacts"
23    }
24  ]
25}
Capabilities

What the data catalog does

  • Every connected system

    Postgres, MySQL, Snowflake, BigQuery, Redshift, Databricks Unity Catalog and HubSpot, down to column types, keys and comments.

  • Who uses this column?

    Each table and column shows the ontology concepts bound to it, so impact analysis before a rename or drop is a lookup, not a guess.

  • Copies linked to originals

    Tables mirrored or federated into another system are linked back to their source, showing where data is duplicated and which copy to query.

  • Metadata only

    Built from catalog metadata and cheap discovery calls. The catalog never reads your rows, so connecting a system is low-risk.

  • Searchable by agents

    AI agents call search_catalog over MCP or the API to find the right table before asking a governed question through the ontology.

  • Governance on Pro

    PII classification and masking, compliance tags, lineage, and write-back of reviewed descriptions and tags to Unity Catalog.

0

Validated templates

0

SQL targets

0

Modeling paradigms

0+

Industries covered

Native output for the warehouses you already run

SnowflakeBigQueryDatabricksRedshiftPostgreSQLSynapseFabricDuckDBdbtSnowflakeBigQueryDatabricksRedshiftPostgreSQLSynapseFabricDuckDBdbt
FAQ

Frequently asked questions

What is a data catalog?
A data catalog (or data catalogue) is an inventory of the data an organization has: which systems exist, which tables and columns they hold, what they mean, who uses them and how sensitive they are. It is how people, and increasingly AI agents, find the right data before they query it.
What does offdataai's data catalog contain?
Every system, table and column from the connections you add: PostgreSQL, MySQL, Snowflake, BigQuery, Amazon Redshift, Databricks Unity Catalog and HubSpot. For each table it shows which ontology concepts are bound to it, so you can answer "who uses this column?" with a lookup, and it links copies of the same data to their originals.
Does the catalog read my data?
No. The catalog is built from metadata only: table and column names, types, keys and comments from the Database Modeler's captured catalogs, plus cheap metadata-only discovery such as listing Unity Catalog tables. Nothing in the catalog reads a row of your data.
How does it find duplicated data?
When the same table exists in two places, for example a Postgres table mirrored into a Databricks catalog through Lakehouse Federation or replication, the catalog links the copy to its original. That tells the semantic layer which copy is cheapest to query, and tells you where the same data lives twice.
Is this a replacement for an enterprise data catalog like Collibra or Alation?
Not necessarily. offdataai's catalog is tied to its data models, ontology and semantic layer: its job is to show what your concepts and metrics are built on and to give AI agents a searchable inventory. If you need enterprise-wide stewardship workflows across hundreds of systems, a dedicated catalog may still fit; many teams start with offdataai because it covers the warehouse and the agent use case without a separate rollout.
How do AI agents use the catalog?
Through the search_catalog tool on offdataai's MCP server, or the public API. An agent can look up tables and columns by name or concept, see which are classified as personal data, and then ask a governed question through the ontology instead of guessing which table to query.
Does the catalog handle PII and governance?
Yes, on Pro. Every column in a model carries a classification and an optional masking strategy (hash, partial or full), and offdataai reports PII inventory, masking coverage and compliance tags such as HIPAA and GDPR, with lineage from source to model. On Databricks it can write reviewed descriptions, keys and PII tags back to Unity Catalog.
What is the difference between a data catalog, a semantic layer and an ontology?
The catalog says what data exists and where. The ontology says what it means: concepts, relations and metrics. The semantic layer uses the ontology to turn questions into correct SQL. offdataai keeps all three connected, so a concept always knows its tables and a table always knows its concepts.
Which plan includes the data catalog?
The data catalog and ontologies start on Starter ($29/month). Live connections, the Database Modeler, governance and lineage, and agent access through the API and MCP server start on Pro ($99/month).
Keep exploring

Explore offdataai

Part of the offdataai platform

From business requirements to production-ready data architecture and AI-ready enterprise context.

See all six stages →
  1. 01DesignLogical, physical and dimensional models.
  2. 02GenerateSQL, dbt, documentation and migrations.
  3. 03DefineBusiness metrics, semantic models and ontology.
  4. 04ConnectReal warehouse schemas and metadata.
  5. 05GovernDrift, lineage, impact analysis and approvals.
  6. 06ActivateMake trusted context available to AI agents.

This page covers Connect and Govern. Every stage works on the same model, so what you design here carries through to the rest.

Start for $1

From requirements to AI-ready context, in one platform.

Describe your business in a conversation, start from a template or connect your own database. Your first model is ready in minutes.

Get startedBook a demo

Plans from $1/month · cancel any time