Built for teams that have to defend their work.
Legal and procurement teams handle some of the most sensitive documents in a company. Here's exactly how Queviny treats that data — no vague assurances, no marketing badges without substance behind them.
Explainable by default
Every redline, extraction, and risk flag ships with a plain-language reason — nothing is automated silently, including at the model level.
Encrypted throughout
Contract data is encrypted in transit with TLS 1.2+ and at rest with AES-256, across every environment that touches it.
EU data residency
Customer data is stored and processed in EU data centers by default, with infrastructure operated from Germany.
Least-privilege access
Role-based access controls, SSO/SAML on Scale plans, and internal access to customer data limited to what support actually requires.
Full audit trail
Every redline, approval, and extracted obligation is timestamped and attributed — exportable for your own audit and dispute needs.
Your data trains your models only
Contract data and playbook rules are never used to train models shared across other customers, and never used for any purpose beyond your account.
How data flows through Queviny
When a contract reaches Queviny — by upload, email intake, or a connected repository — it's encrypted in transit and stored encrypted at rest in an EU data center. The review model reads the document to extract clauses, compare them against your playbook, and identify obligations; none of that processing leaves Queviny's environment or gets shared with a third-party model provider without a data processing agreement in place.
Redlines, extracted obligations, and risk flags are written back to your account's data store, visible only to users your organization has granted access to. Nothing is sent externally — to a counterparty, to e-signature, to anywhere — without an explicit action from your team.
Playbook rules and correction history — the record of when your team accepts, adjusts, or overrides a suggestion — are used exclusively to sharpen Queviny's understanding of your account. They are never pooled into a shared training set or used to improve results for any other customer.
Common questions
Where is our contract data actually stored?
In EU-based data centers by default. Enterprise customers with specific residency requirements can discuss regional options as part of a Scale plan.
Do you train shared models on customer contracts?
No. Playbook rules and contract data are used only to improve the model's understanding of your own account — never pooled with, or used to train models serving, other customers.
What compliance certifications do you have?
We operate under a SOC 2 Type II control framework, currently in the audit process, and are built around GDPR requirements as a German-based company handling EU customer data by default. Full audit documentation is available under NDA for Scale plan evaluations.
How is access to our contracts controlled internally?
Role-based permissions scope every user to what they need, and internal Queviny staff access is limited to what's required for active support requests, logged, and time-boxed.
Can we delete our data if we leave?
Yes. On request, we delete customer data — including contracts, extracted obligations, and playbook configuration — from production systems within 30 days, and from backups on their standard rotation schedule.
What happens if a redline or flag turns out to be wrong?
Queviny proposes; your team decides. Nothing is sent, approved, or rejected automatically without a configured rule your team set — which is also why every suggestion ships with a reason your reviewers can check.
Security or compliance questions we haven't answered here?
Reach out directly — we're glad to walk through our architecture, data flow, or audit documentation with your team.
14-day free trial · No credit card required