Patchwork, not federal
California, Colorado, and others have their own privacy statutes. The deployment documents the data path each of them can examine.
Self-hosted AI / the United States
Self-hosted AI for US companies, built for a patchwork of rules.
Private AI deployed for US businesses with the document store on your accounts and hosting in a US region. There is no single federal data protection law in the US, so the deployment documents where data goes, who processes it, and what state or sector rules apply — before anything is loaded.
From $3,500 / ₦2.5m floorus-east-1 / us-west-2 optionsClient-owned accountsFull handover
Example only / architecture varies by client
Designed for review against state privacy laws and professional confidentiality rulesa US region (us-east-1 Virginia or us-west-2 Oregon) optionClient-owned billingFull handover
Why it matters here
The US has no single privacy law. It has rules that multiply.
State statutes, sector rules, and professional confidentiality duties all apply at once. The deployment is built to be reviewable against each.
HIPAA and the rest
Healthcare and other sectors add their own processing rules. The architecture names every provider and channel so sector review is possible.
Confidentiality is a rule
For lawyers, ABA Model Rule 1.6 requires reasonable efforts to protect client information. The design is built to support that effort.
The three levels
The architecture follows the constraint.
There is no single correct answer. Quality, privacy, residency, and budget decide the path.
Your cloud, in the US
App and document store in your account in a US region (us-east-1 Virginia or us-west-2 Oregon), with the model on your key. No GPU required.
For location requirements
An approved cloud region or managed service when residency and procurement matter.
For strict control
Model, embeddings, and retrieval on your hardware when external inference is not acceptable.
What it really takes
Measured workloads, not GPU shopping lists.
Hardware is sized after the workload is real. These are starting points, not quotes.
4 to 8 vCPU, 16 to 32 GB RAM
No GPU required for the app and document store. Model usage bills stay on the client account.
From 24 GB VRAM
A starting point for a small quantized text model. Model size, concurrency, and latency decide the real requirement.
Power, cooling, updates, backups
A GPU is one part of the environment. A partner data centre can be more realistic than an office server room.
Not the first promise
We size hardware after the workload is proven. No GPU inventory before a signed scope and deposit.
Straight answers
Common questions.
Where would US client data be hosted?
A US region such as us-east-1 (Virginia) or us-west-2 (Oregon), documented in the architecture. Fully local deployment is the strict option.
Which US privacy law applies to us?
It depends on your state, your sector, and your clients. That is exactly why the deployment documents the data path first: the relevant rules can then be reviewed against it.
Does this support HIPAA-covered work?
The architecture can be scoped so no external model provider receives protected health information, with the data path documented for the client's compliance team to review.
Do invoices come in US dollars?
Yes, from $3,500 for a first workflow, with the naira floor unchanged.
In the US, the rule book is a patchwork. The data path is the thread that holds it together.
Why trust Pristine3D?
We build and operate production software.
Pristine3D Ltd builds and operates live digital products, and we run private AI workflows internally as part of our own operations. We scope around your real workflow: the documents you own, the questions your team asks, and the access boundary you approve. Based in Lagos, Nigeria, we work remotely with clients worldwide.
We already run live products
Based in Lagos, Nigeria, Pristine3D currently builds and operates smartcards.ng, venu.ng, photoshoot.ng, and ugc.ng in production.
We start with the actual workflow
One input, one output, one test set, and one person who owns the result. We scope a real workflow instead of a transformation programme.
The boundary stays visible
Cloud, model, storage, and messaging accounts stay in your name. The chosen data path, access rules, test record, documentation, and training are part of the agreed scope.
Pricing / fixed scope
Know the starting numbers before you ask.
The final quote follows the workflow. Infrastructure and model bills stay on your accounts.
One Workflow
Starting from$3,500 / ₦2.5m
2 to 3 weeksOne workflow for a small team, with training and handover.
See the one-workflow packageFirm-wide
Starting from$7,500 / ₦8m
4 to 6 weeksMultiple sources, roles, integrations, and admin handover.
See the deploymentAnnual support
Starting from$3,000 / ₦1.5m
per yearStandard care for one delivered workflow. Optional. Larger deployments and active monitoring are separately scoped.
See supportArchitecture review from $500. Standard annual support is $3,000 / ₦1.5m per year for one delivered workflow. New workflows, integrations, active monitoring, and infrastructure are separately scoped; infrastructure, model, storage, and messaging bills stay on client accounts. Full pricing and what changes the quote
Own your knowledge base
The model is not the product. The knowledge base is.
Documents, the retrieval index, access rules, and the workflows built around them are the asset, and they compound. We deploy so the knowledge base stays yours: on your accounts, in the environment you choose, under access rules your team defines. The model behind the answers is a connector, so the knowledge base moves with you, not with a vendor.
Your corpus, your index
The document store, metadata, and retrieval setup live on accounts you own. No vendor holds the corpus.
Models are swappable parts
Change the model provider, move regions, or go local without rebuilding the knowledge base or the workflow.
The knowledge base is the alpha
Every improvement to the corpus improves the answers, and the improvement stays with you, not with a vendor.
Keep exploring
Related setups.
Start with the constraint
What is your constraint?
State rules, sector requirements, procurement, or a no-external-inference rule. Describe it and we will map the architecture.