offdata ai — agentic AI for data modelers and data engineers

Google BigQuery

AI BigQuery schema generator

Describe your domain in plain English. Ship a BigQuery warehouse — DDL with PARTITION BY and CLUSTER BY, plus a complete dbt project with sources, staging, marts, tests, and seeds.

ddl/bigquery.sqlbigquery sql
CREATE TABLE analytics.dim_customer (
  customer_sk   STRING       NOT NULL,
  customer_id   STRING       NOT NULL,
  name          STRING,
  plan_id       STRING,
  valid_from    DATE         NOT NULL,
  valid_to      DATE
)
PARTITION BY valid_from
CLUSTER BY customer_id, plan_id
OPTIONS (description = "SCD2 customer dimension");

CREATE TABLE analytics.fct_usage (
  usage_id      STRING       NOT NULL,
  customer_sk   STRING       NOT NULL,
  plan_id       STRING       NOT NULL,
  event_ts      TIMESTAMP    NOT NULL,
  api_calls     INT64        NOT NULL
)
PARTITION BY DATE(event_ts)
CLUSTER BY customer_sk, plan_id;

OffDataAI is purpose-built for BigQuery. Our generator emits BigQuery-native DDL with the right types (TIMESTAMP, DATE, NUMERIC(p,s), STRUCT, ARRAY), correct PARTITION BY (typically on event date), and CLUSTER BY on the columns you actually filter on. From the same conversation, you also get a complete dbt project ready for dbt build.

What OffDataAI generates

  • BigQuery-native DDL

    PARTITION BY DATE(event_ts), CLUSTER BY, NUMERIC(p,s), STRUCT and ARRAY, integer-range partitioning — never generic ANSI SQL.

  • Kimball, Data Vault 2.0, or 3NF

    Three paradigms supported. Pick the one your team uses; the generator emits matching partitioning and clustering.

  • Full dbt project

    dbt_project.yml, sources, staging, marts (with partition_by + cluster_by), schema tests, and seeds. dbt build works on the first run.

  • Cost-aware design

    Partitions and clustering are chosen to minimize scan cost on the queries you described — facts partition by event date, dims cluster by natural key.

Frequently asked questions

Can OffDataAI generate BigQuery DDL from plain English?
Yes. Describe your business domain in plain English. OffDataAI's interview agent gathers grain, cardinality, and partitioning hints, synthesizes a typed Intermediate Representation, and emits BigQuery-native DDL with PARTITION BY and CLUSTER BY.
Does OffDataAI use BigQuery-specific features like PARTITION BY?
Yes. OffDataAI's BigQuery generator emits PARTITION BY clauses on facts (typically by daily ingestion or event date) and CLUSTER BY on the natural-key columns most often filtered. Types use BigQuery-native TIMESTAMP, DATE, NUMERIC(p,s), and STRUCT — never generic ANSI types.
Does it support Kimball, Data Vault, and 3NF on BigQuery?
Yes. All three paradigms are first-class for BigQuery. The synthesis agent shapes the IR; the BigQuery generator emits the right partitioning and clustering for each paradigm.
Does it generate a dbt project for BigQuery?
Yes. OffDataAI emits a complete dbt project for BigQuery — dbt_project.yml, sources, staging, marts (with partition_by and cluster_by), schema tests, and seed CSVs.
How is this different from asking ChatGPT for BigQuery DDL?
ChatGPT can sketch DDL, but often confuses BigQuery PARTITION BY with Snowflake CLUSTER BY or PostgreSQL partitioning. OffDataAI runs a dedicated BigQuery generator after a validated IR — the output is reproducible and warehouse-correct.

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.