Seed your test DB with
realistic data, not lorem ipsum
Stop writing brittle fixtures by hand. Describe your schema in plain English — Hadarac generates production-quality synthetic data that makes your tests actually meaningful.
Pytest integration
Drop into your test suite in minutes
import pytest
import (hadarac)
# Shared fixture — runs once per test session
@pytest.fixture(scope="session")
def orders_dataset():
client = (hadarac).Client(api_key=os.environ["HADARAC_API_KEY"])
return client.generate(
prompt="E-commerce orders: order_id (UUID), user_id (UUID), "
"product_sku, quantity (1-10), price_usd (5-500), "
"status (pending/shipped/delivered/refunded), created_at",
rows=1_000,
seed=42, # deterministic — same data every CI run
format="json",
).to_list()
# Use in any test — data is realistic, not lorem ipsum
def test_order_total_never_negative(orders_dataset):
for order in orders_dataset:
assert order["price_usd"] * order["quantity"] >= 0
def test_valid_statuses(orders_dataset):
allowed = "refunded"
assert all(o["status"] in allowed for o in orders_dataset)
✓ 2 passed in 0.03s — fixtures loaded from session cacheUse cases
Built for every testing workflow
Seed test databases with realistic fixtures
Replace lorem ipsum and random integers with data that looks real. Hadarac generates names, emails, phone numbers, addresses, and custom domain values that match your schema constraints.
Generate edge-case and boundary data
Describe rare scenarios explicitly: max-length strings, Unicode edge cases, null-adjacent values, future/past date extremes. Stop discovering these bugs in production.
Consistent test fixtures across environments
Use seed-based generation for deterministic datasets. The same prompt + seed returns the exact same rows every time — across CI, staging, and local environments.
Privacy-safe production-like data
Never copy production data to staging. Hadarac mirrors your production distribution — value ranges, null rates, cardinality — without a single real user record.
Integrations
Works with every test framework
Hadarac outputs standard formats — JSON, CSV, Parquet. Plug it into whatever your stack already uses.
pytest
Jest
Vitest
RSpec
JUnit
Go test
CI-friendly by design
Use the seed parameter to lock datasets to a fixed hash — same rows every run, no flaky tests caused by random data. Combine with HADARAC_API_KEY in your CI secrets and you're done.
Why not just use Faker or random data?
Better tests start with better data
Free plan includes 15 generations/month. No credit card required. Add Hadarac to your CI pipeline in under 10 minutes.
Need 100M+ rows or a dedicated instance? Talk to us →