Your data is yours. The jurimetrics is everyone's.
Two layers, two rules. Your company's sensitive data is sovereign and isolated; jurimetrics comes from public labor-court records, collective and with no claimant identified. The two coexist without one touching the other.
Personnel records · documents · defense strategy
Labor-court decisions · aggregated · no claimant identified
Private, isolated, only yours.
Personnel records, documents and your defense strategy. They stay in your dedicated environment, never crossing with another client, never training models. But isolated isn't idle: your evidence on each claimant is what refines your case's probability, yours alone, never flowing back to the collective layer.
- —Dedicated environment per client
- —No model training on your content
- —Your evidence sharpens your number, inside your environment
- —Named access by role and authentication records
Public, collective, open records.
Public labor-court decisions, aggregated. Lawyers and judges appear because they act publicly in the case files; claimants are not identified in this layer.
- —Grant pattern by topic, court and judge
- —Patterns in how lawyers and judges act
- —All from public records, with no claimant identified
The boundary is strict: the collective layer uses only public records; your sensitive data never leaves your environment.
No training on your data
Your documents and cases do not train AI models, and are not used for advertising or resale. Content sent to the provider is not retained beyond the time needed to answer.
Recommendations anchored to the source
Every recommendation carries the excerpt and the document it rests on, so it can be checked. The reading is informational and statistical: the decision is the lawyer's, and the court's official system prevails.
Data protection and governance
We act as a processor, on your company's instructions. Dedicated environment per client, named access by role, authentication records, and encryption in transit and at rest.
Security in five questions
If my data stays isolated, does it help the analysis?
It does, and that's where the value is. The collective layer comes from public records, with no claimant identified; your evidence on the claimant (almost always a former employee) adjusts your case's probability, inside your environment. The reverse never happens: your data does not feed the collective layer or train models.
Do you use my company's data to help other clients?
No. Collective jurimetrics comes from public court decisions. Your personnel records, documents and strategy stay in your dedicated environment and do not feed the base made available to another company.
Then how does cross-company benchmarking work?
On public records. Lawyers and judges act publicly in the case files, and cross-referencing those patterns is reading public sources, not your data. Claimants are not identified in the collective layer.
Do you train models on my documents?
No. Your documents are not used to train models, by us or by the providers acting on our behalf, who are bound by contract to the same obligation.
How do you comply with LGPD?
We act as a processor, on your company's instructions, and your company is the controller. Dedicated environment per client, named access by role, authentication records, and encryption in transit and at rest. The detail is in the Platform Privacy Notice.
Public documents: Privacy Policy · Platform Privacy Notice (in Portuguese)
Walk through security with us.
We show the two-data boundary and the security boundary in practice.
Book a demo