SIMDATA

SYNTHETIC RECORDS API

The same person
in every table.

You keep a dictionary of how fields relate. You pass a seed. Tomorrow’s person is still today’s person — name, IDs, and the visit that belongs to them.

869,098 synthetic records generated

Create free account Read the docs

Starter credit is already on the account. No card.

curl -H "Authorization: Bearer $SIMHEALTH_API_KEY" \
  -H "Accept: application/json" \
  "https://simdata.interoperabilitypro.com/v1/random/person?count=1&seed=1842&gender=female"

Same seed Same crowd tomorrow

One request Up to 10 million records

Your rules Edit lists, not code

Safe fixtures Never leak real user data

How it works

  1. 01

    Write the rules

    Lists of values, how often they show up, and how records nest. Change them in the app. You don’t ship a new binary.

  2. 02

    Ask for a draw

    Pass a seed to get the same records back. Add gender=female when you only need that slice.

  3. 03

    Take a file

    JSON when you want to read it (10,000 max). Avro when you’re filling a pipeline. Ten million rows is a normal Tuesday.

That’s the loop. Create a free account and we’ll give you a key.

API

https://simdata.interoperabilitypro.com · sign in, then send a Bearer key · /llms.txt

GET /v1/random/{recordType}
How many, which seed, optional filters. Avro unless you ask for JSON.
GET /v1/catalog
What’s in your dictionary.
GET /v1/catalog/{kind}/{id}
One entry, including the source you edit.
PUT /v1/catalog/{kind}/{id}
Replace the source. If it doesn’t compile, nothing goes live.
GET /v1/stats/records
Running total. No key needed.

Generate records · Catalog API

Design philosophy

Coherence

If her name is Maria, the linked record is Maria’s. IDs don’t drift between tables.

Scale

Ten million rows should feel ordinary, not like a sales call.

Mix

A crowd, not one row pasted a thousand times.

Yours

Change a list. The next call uses it.

Questions or customer support?

We read every message at support@interoperabilitypro.com.

Email support