About
We're building the data
infrastructure for the AI era
AI teams need massive, realistic, privacy-safe datasets. We built Hadarac so they can generate them instantly — without the legal risk, the engineering overhead, or the weeks of waiting.
Origin story
Why we built this
We were building ML pipelines and kept hitting the same wall: the real data we needed was either too sensitive to use, too sparse to train on, or locked behind months of legal review. We'd spend more time wrangling data than building models.
The existing synthetic data tools were either academic toys — too rigid, too slow — or enterprise products that cost $200k and required a 6-month procurement process. There was nothing in between for the team moving fast.
We built Hadarac to fill that gap. Plain-English generation. Instant results. Privacy built in. A tool that treats data as a first-class product, not an afterthought.
Principles
How we work
Speed is a feature
Every second a data team waits for a dataset is a second they can't build. We optimise for instant — instant generation, instant results, instant iteration.
Privacy by design
We don't bolt privacy on at the end. Zero retention, realistic PII replacement, and compliance-ready outputs are core to everything we ship.
Quality over quantity
One honest, well-crafted feature beats five half-finished ones. We ship things we'd want to use ourselves — and we sweat the details.
Customer obsession
We read every support ticket. We talk to users every week. The product roadmap is built from real pain, not internal assumptions.
Backed by
Placeholder logos — update when fundraise is announced.
Join the team
We're a small, focused team building something important. If that sounds like your kind of challenge, we'd love to hear from you.
See open roles