Trust is architecture, not a slogan
Among its founders is a lawyer — and the protection of firms’ and clients’ data is built into the foundations, not added on top of them. What follows describes what we have actually built and what we test continuously; it is not a list of intentions.
Complete isolation between firms
Isolation at the level of the database itself (Row-Level Security, enforced) — not at the level of the code alone. Proven by more than 500 automated tests, including deliberate attempts to break it.
Your data trains no one
No model training on client data — an architectural rule that is not open to negotiation, not a setting to be switched.
Every action is recorded
Audit records that are never deleted: who did what and when — including our own access to any file.
Declared limits
What the system knows and what it does not is shown to the user — and “we don’t know” is an answer we respect rather than replace with a guess.
Questions people actually ask
Is my data safe?
Each firm’s data is isolated from every other firm’s within the database itself, so no firm reaches a single row belonging to another even if the code goes wrong. The isolation is tested automatically with every change, not once at delivery.
Does anyone on your side look at my clients’ files?
Access is an act recorded in the audit log with the name of whoever accessed it and the time — including our own access for technical support. The log is never deleted, so what happened stays readable to you.
Where is the privacy notice?
It is under legal review and has not been published yet. We will not put up a temporary text: a published privacy notice is a commitment the company signs, and an unapproved text is worse than its absence. It will be published the day our founding lawyer approves it.
A question you did not find answered?
Write to us directly — we answer with what we have actually built, and what we have not built we say we have not built.