A target discovery program is the whole company.
Don't run it through someone else's model.
Private LLM hosting on a dedicated machine in EU-Central, for early-stage biotech and life sciences R&D teams working with lab notebooks, sequence data, and target discovery IP that shouldn't sit on shared AI infrastructure.
Not legal, regulatory, or investment advice. This page describes GPUwerk's infrastructure and contractual commitments only. It's aimed at early-stage R&D: lab notebooks, target discovery, and assay data, not clinical trial participant data or GxP manufacturing records. For that framing, see our pharma and life sciences page instead. Whether a given use of your research data is permitted under your own collaboration agreements, IP assignments, and applicable regulation is a determination for your legal and scientific leadership.
The one asset a small team has, one shared answer
An early-stage biotech often has one real asset before it has revenue: the discovery data itself. A leaked screening result or a competitor seeing a target thesis six months early can change a company's fundraising outcome. That raises the same infrastructure question before anything else: where does the model actually run.
"Our lab notebooks and screening data are the entire cap table's bet"
Unpublished target hypotheses, assay results, and sequence data aren't protected by statute the way trial participant data is, which makes a shared AI vendor's retention policy or training pipeline a pure self-inflicted risk to the company's valuation. A dedicated single-tenant machine means no other tenant's workload, and no vendor model training, is anywhere near that dataset.
"A collaboration agreement with a pharma partner restricts where our data can go"
Licensing and co-development deals commonly specify data-handling terms your team has to be able to answer for. On a GPUwerk instance, the machine is assigned to your organisation alone, in EU-Central. GPUwerk operates the machine but, under the data processing agreement, does not access your content except at your request for support or where a legal obligation requires it.
"Scientists already use consumer AI tools to draft grant sections and summarise papers"
Drafting a grant narrative or summarising a preprint in a public AI tool is common, and rarely reviewed for what unpublished result went in alongside it. A sanctioned alternative on your own instance, same chat interface, keeps that workflow off third-party infrastructure. See the private ChatGPT setup →
What's actually in the contract
For your head of R&D, IT, or legal counsel to review directly. This is a description of infrastructure, not a regulatory or IP determination.
| Question | Answer |
|---|---|
| Where does it run? | EU-Central, on a dedicated single-tenant machine assigned to your organisation |
| Who operates it? | PRINT IT! SE, a Societas Europaea registered in Prague, Czech Republic. No US parent entity. |
| Who can reach the instance? | Through the instance itself, only holders of your SSH keys; password login is disabled fleet-wide. GPUwerk keeps infrastructure administrator access to the underlying machine, as on any hosted service, and under the DPA does not use it on your content except at your request for support or where a legal obligation requires it. |
| Is this the right page for clinical trial or GxP data? | No, see the pharma and life sciences page for that framing; the infrastructure is the same, the compliance discussion differs. |
| Does GPUwerk read lab notebooks or sequence data? | No. We host the hardware and, under the DPA, do not access, read, copy, index or analyse workload content. |
| Sub-processors for the workload? | None, listed at /legal/sub-processors |
| DPA (GDPR Art. 28)? | Published at /legal/dpa, no charge |
| Data on termination? | Container and workspace volume deleted from the node, then the node is sanitised before reassignment. Filesystem deletion, not a cryptographic erase; export what you need before terminating, since it isn't reversible. |
Questions we get from biotech R&D teams
How is this different from GPUwerk's pharma page?
The pharma page addresses clinical trial data and GxP-adjacent manufacturing and compliance concerns. This page is aimed earlier in the pipeline: early-stage biotech R&D, where the sensitive material is target discovery data, lab notebooks, and sequence or assay data rather than trial participant records or manufacturing records. The underlying infrastructure answer is the same dedicated single-tenant machine in EU-Central; see the pharma page for the clinical and manufacturing framing.
Does GPUwerk train on our sequence data or lab notebooks?
No. GPUwerk operates infrastructure, not models. Whatever you run on your instance is your software and your data, and under our data processing agreement GPUwerk does not access that content except at your request for support or where legally required.
Is this suitable for data covered by a licensing or collaboration agreement with a pharma partner?
We can describe the infrastructure: dedicated single-tenant machine, EU-Central, an Article 28 DPA, no sub-processors on the workload. Whether that satisfies a specific collaboration agreement's data-handling clause is a determination for your legal team and your partner's, not something GPUwerk can certify.
Do you sign a DPA?
Yes, a standard GDPR Article 28 DPA is published at /legal/dpa at no charge, and there are no sub-processors for instance workloads.
Related pages
Put the discovery data on hardware your own team controls.
Talk to the people who run the racks, or start with a pilot and a practical rollout plan.
Talk to us Deploy an instance