offdata ai — agentic AI for data modelers and data engineers

Data Vault 2.0

AI Data Vault 2.0 generator

Describe your domain in plain English and OffDataAI builds the Data Vault for you — hubs, links, and satellites with hash keys and hash diffs — then ships native DDL and a dbt project. Standards-compliant Data Vault 2.0 without hand-modeling every structure.

raw_vault/customer.sqldata vault · ddl
-- Hub: business key
CREATE TABLE raw_vault.hub_customer (
  customer_hk      BINARY(20)    NOT NULL,  -- hash key
  customer_bk      STRING        NOT NULL,  -- business key
  load_dts         TIMESTAMP_NTZ NOT NULL,
  record_source    STRING        NOT NULL,
  CONSTRAINT pk_hub_customer PRIMARY KEY (customer_hk)
);

-- Link: relationship between hubs
CREATE TABLE raw_vault.link_customer_order (
  customer_order_hk BINARY(20)   NOT NULL,
  customer_hk       BINARY(20)   NOT NULL,
  order_hk          BINARY(20)   NOT NULL,
  load_dts          TIMESTAMP_NTZ NOT NULL,
  record_source     STRING        NOT NULL
);

-- Satellite: descriptive, time-variant context
CREATE TABLE raw_vault.sat_customer_details (
  customer_hk      BINARY(20)    NOT NULL,
  load_dts         TIMESTAMP_NTZ NOT NULL,
  hash_diff        BINARY(20)    NOT NULL,  -- change detection
  email            STRING,
  plan_id          STRING,
  record_source    STRING        NOT NULL,
  CONSTRAINT pk_sat_customer PRIMARY KEY (customer_hk, load_dts)
);

Data Vault 2.0 is powerful but tedious to model by hand: every business key becomes a hub, every relationship a link, every descriptive context a satellite, and each needs consistent hash keys and change-detection logic. OffDataAI automates it. Pick the Data Vault paradigm, describe your business, and the synthesis agent separates business keys, relationships, and descriptive attributes into a standards-compliant raw vault. Hash keys (_HK) and HASH_DIFF columns are generated by the DDL layer — never hand-added — so loads are deterministic across Snowflake, BigQuery, Databricks, Redshift, Postgres, Synapse, and Fabric.

What OffDataAI generates

  • Hubs, links, and satellites

    Business keys become hubs, relationships become links, and time-variant context becomes satellites — automatically separated from your domain description.

  • Hash keys and hash diffs

    _HK hash keys and HASH_DIFF change-detection columns are emitted by the DDL layer for deterministic, parallelizable loads — you never write them by hand.

  • Insert-only and auditable

    Generated structures follow Data Vault's insert-only, fully-auditable pattern with load timestamps and record sources on every table.

  • Extends without breaking

    New sources or attributes add satellites and links instead of altering existing tables — and OffDataAI generates the migration DDL for each change.

Frequently asked questions

Can OffDataAI generate a Data Vault 2.0 model?
Yes. Choose the Data Vault 2.0 paradigm and describe your domain. OffDataAI identifies business keys and relationships and produces hubs (business keys), links (relationships), and satellites (descriptive, time-variant attributes) — with hash keys and hash diffs generated by the DDL layer. You get native DDL plus a dbt project to load it.
Does it generate hash keys and hash diffs?
Yes. Hash keys (_HK) on hubs and links and HASH_DIFF columns on satellites are generated automatically by the DDL layer — you never hand-write them. They're produced consistently across every target platform so loads are deterministic.
Which platforms can the Data Vault target?
Snowflake, BigQuery, Databricks, Redshift, Postgres, Synapse, and Fabric. The hub/link/satellite structures compile to each platform's native DDL with the right types, clustering, and distribution.
Can I evolve the Data Vault as the business changes?
Yes. Add a new source, attribute, or relationship and OffDataAI extends the model with new satellites or links rather than breaking existing structures — Data Vault's core advantage. Each change is a new version, and OffDataAI can generate the migration DDL.
Do I need to be a Data Vault expert to use it?
No. You describe the business; OffDataAI applies Data Vault 2.0 patterns — separating business keys, relationships, and descriptive context — so you get a standards-compliant model even if you're new to the methodology. Experts can inspect and adjust every structure.

Your data warehouse is one conversation away.

Describe your domain, or open one of 150+ production-grade templates. ERDs, DDL, and a complete dbt project — generated in under a minute.