Relational synthetic data · for engineering teams

Fake data that keeps your foreign keys honest.

Testing with production data is a GDPR breach waiting to happen. Hand-crafting consistent fakes takes hours. Paste a schema — get millions of realistic, relationally-intact rows in seconds.

Get started →See how it works
synthdata-flow.sql
-- topologically ordered · FK-safe INSERT INTO "User" (id, email) VALUES (1, 'mae.fournier@hotmail.fr'), (2, 'leo.bianchi@gmail.com'); INSERT INTO "Order" (id, user_id) VALUES (1, 2), -- → User(2) ✓ (2, 1); -- → User(1) ✓
The pipeline

From schema to seed in three moves.

No agents to install, no access to your production database required.

01

Paste your schema

Drop in your PostgreSQL or MySQL CREATE TABLE statements — or MongoDB $jsonSchema validators. We parse columns, constraints and references into a relational graph.

02

We respect the graph

A topological sort generates parents before children, so every foreign key points to a row that actually exists. Unique columns stay unique.

03

Export & seed

Download a ready-to-replay SQL dump, per-table CSVs or NDJSON for mongoimport. Free of real names, emails and card numbers — GDPR-safe by construction.