Castellan · Heritage
Castellan · Heritage

Built on the Bitkern Group.

Castellan does not start from zero. It draws on a decade of experience at the Bitkern Group — developing and running energy-intensive compute infrastructure in Switzerland, Austria and the United States — and starts with access to the one resource that decides every AI project before the first rack arrives: power.

10yrs
Of operating experience the group brings to Castellan
+300MW
Built and operated by the group to date
+5GW
Of power accessible through our network
Energy
Established power-supply & procurement relationships
2017 → 2032 Drag to explore
2017
Bitkern Group founded
Zug, Switzerland. Crypto-mining and high-performance data centres.
2017 – 2026
+300 MW built & operated
A decade of power procurement, high-density build-out and 24/7 operations across Switzerland, Austria and the United States.
2026
Castellan Intelligence
Independent Swiss company founded to develop, build and lease AI data centres — on its own, purpose-built sites.
2026
+5 GW power access
Power available through our supplier and site network, as demand requires.
2027
Phase I construction
Greenfield campus of at least 50 MW breaks ground in Europe or North America.
2028
Phase I online · 50 MW
Our first campus is commissioned — leased as powered shell or turnkey data centre on triple-net terms.
2030
200 MW cumulative
Phase II adds a second campus.
2031
400 MW cumulative
Phase III — a mature, multi-site operating platform.
2032
800 MW platform
At least 800 MW of AI capacity, scalable as demand requires.
01
The Bitkern story

A decade of energy-intensive compute.

The Bitkern Group has spent a decade at the demanding end of the compute business — crypto-mining and high-performance data centres: securing power, building high-density facilities and operating them commercially, around the clock, in Switzerland, Austria and the United States. Many of today's AI-infrastructure operators came out of the same discipline.

Over that decade the group has built and operated more than 300 MW of energy-intensive compute capacity — negotiating multi-year power supply, engineering high-density electrical and cooling systems, and running facilities where every hour of downtime carries a price. That is a different education from the real-estate playbook most data-centre entrants arrive with. It teaches what breaks at scale, what a grid operator will actually commit to, and what a megawatt really costs across a full market cycle.

It also built the network that gives Castellan access to more than 5 GW of power, available as demand requires, together with established energy-supply and procurement relationships. The hardest part of an AI campus is not constructing the building — it is securing abundant, affordable, interconnected power. That is why Powered Land sits at the base of everything we offer.

Most AI-infrastructure entrants start from zero on power. Castellan starts with access already in place.

02
Operating capabilities

Six capabilities behind every project.

These are not aspirations on a slide. They are disciplines proven over a decade at the Bitkern Group — from energy sourcing to GPU Procurement & Trading and HPC Hosting — and brought into Castellan by the people who practised them, together with their supplier network.

01
Powered Land & energy sourcing
Access to more than 5 GW of power, power-zone identification, interconnection studies and multi-year supply relationships built over a decade of energy-intensive operations. Power is secured before a shovel hits the ground.
02
Data-centre engineering
Electrical topology, transformer planning, cooling design — air, evaporative and direct-liquid-cooling-ready — and structured cabling backbone, engineered for high-density GPU deployments from H100/H200 to Blackwell-class, built to data-centre standards. Delivered as powered shell or full turnkey fit-out.
03
Swiss counterparty & governance
A Swiss stock corporation in Zug as your contractual counterparty, with local project entities where a site requires them — structured so each campus answers to the jurisdiction its tenants need.
04
Built for regulated workloads
Lease and service documentation prepared with the outsourcing, data-residency and EU AI Act questions of regulated tenants in mind.
05
GPU sourcing & supply chain
The Bitkern Group's supplier network for Blackwell-class GPU systems, networking fabric and storage. Price benchmarking, logistics and commissioning for our customers' hardware — procurement run as an operating discipline.
06
Compute placement & monetization
A decade of running energy-intensive compute commercially: placing workloads and structuring offtake contracts that connect hardware owners with compute buyers. We arrange the capacity; we do not operate GPU fleets of our own.
03
From Bitkern to Castellan

Why Castellan does not start from zero.

Castellan was founded to apply this experience to a new task: developing, building and leasing AI data centres at campus scale, as set out in our vision. What it draws from the Bitkern Group is concrete — four advantages that usually take a decade to build, available from day one.

  • 01
    Power access
    More than 5 GW of power, available as demand requires — the scarcest input in AI infrastructure, reached through relationships that took ten years to build and cannot be replicated on a project timeline.
  • 02
    Supply relationships
    Energy contracts, OEM and distribution lines for GPU systems, and construction and engineering partners — counterparties that already know how our team contracts, prices and delivers.
  • 03
    The team
    Operators, engineers and commercial leads who ran the Bitkern Group's facilities are part of Castellan's team — the leadership team brings that operating record personally, not by acquisition.
  • 04
    Track record
    Ten years of delivery and more than 300 MW built and operated give tenants and their regulators what no newcomer can offer: evidence that the numbers on the page have been achieved before.

Two companies, deliberately separate. Bitkern continues to run its own business, unchanged. Castellan is an independent Swiss company that acquires and develops its own sites — greenfield locations selected and purpose-built for AI, to data-centre standards. No mining facility is converted into an AI campus; the two asset bases do not mix. What Castellan draws from Bitkern is expertise, the supplier and site network — including locations that are unsuitable for mining but well suited to AI — and part of the team.

The result: when Castellan signs its first lease, it does so with power access in hand, a supply chain that has already delivered, and a team that has commissioned and operated this class of infrastructure before. That is the difference between a development story and an operating company.

Judge us by what we have already built.

A decade of operations, more than 300 MW delivered and access to more than 5 GW of power — now applied to data centres for AI. Talk to the team, or see our two delivery models: powered shell and turnkey.