About Suvard

The guardian company.

We build the layer that lets regulated organisations use AI while what must never leave the building stays home, and what happened can be proven afterwards.

The name

Two heritages. One word.

守
suKorean: to guard, as in suho (수호), guardian
ᛦ
vardNorse: the guardian root of vörðr, warden and cairn

su + vard = Suvard: guardian, in both.

The founder's Korean and Nordic roots share one word: the guardian. So does the company.

Why a guardian

Every civilisation built guardians around what mattered most.

AthensNomophylakesGuardians of the laws: officials whose only task was to watch over the law itself.
SpartaThe EphorsFive overseers who watched the kings and guarded the constitution.
SeoulSumunjangThe gate guards of Gyeongbokgung, still changing watch to this day.
The NorthVörðrThe warden spirit of Norse folklore that walks beside you and keeps watch.

We built one for the AI age.

The founder

Matias Park, founder of Suvard

Matias Park

Founder · Danish-Korean, Aalborg

Cybersecurity engineer (IT security, UCN) with a background in healthcare and sports science: the clinical domain knowledge behind the first product. Built CareProxy alone, had the engine validated in Treat Systems' technical environment, is in dialogue with a Danish region, and won the life-tech category at the Danish Championship in Entrepreneurship 2026. Now building the guardian for every regulated sector, and the team to build it with.

The first guardian

CareProxy is the healthcare edition of the engine.

Danish clinical detection, everything on your own network, a signed log where changes can be detected. Validated in Treat Systems' technical environment. Everything Suvard later brings to energy, finance, legal, defence and the public sector is proven here first.

Visit careproxy.dk ↗

Built to support

GDPRNIS2EU AI ActMDR (healthcare)DORA (finance)

Suvard provides the technical controls and the evidence. The legal assessment of whether you comply remains yours. Not legal advice.

Next step

A 30-minute check on where AI is already running in your organization.

We start with a mapping, not a sale. You get an honest picture of what passes through your network, and what it would take to be able to prove it afterwards.