- Home
- Security
Security & trust
Our automations touch financial records, customer data and core systems. Here’s exactly how we protect them.
Principles
Four rules on every project
NDA before anything else
Signed before we see a single file or system.
Runs in your environment
Your servers or your approved cloud account, not ours.
Least-privilege access
Only the access a task needs, revoked at handover.
Every run is logged
See exactly what ran, when, and on what data.
In practice
How a project is secured
- An NDA is signed before discovery begins.
- We request read-only access first, and write access only where the automation needs it.
- Credentials are stored in your secret manager or environment, never in code or chat.
- Every automated run is logged with what it read, what it changed and when.
- AI agents get the narrowest permissions that let them do the job, and every action they take is recorded.
- Inputs are validated; when something looks wrong, the automation stops and alerts a person.
- At handover you receive the code, documentation and all credentials, and we revoke our access.
Compliance
Built for regulated teams
We design with India’s Digital Personal Data Protection Act and GDPR principles in mind: collect only what’s needed, keep it where you control it, and make it easy to delete.
Do you sign an NDA?
Yes, on every engagement, before we see any data or systems.
Where does our data go?
Automations run in your environment or a cloud account you approve. We don’t copy your data to our own servers.
Do you use AI models on our data?
Only where the project needs it, and only with providers and settings you approve. For sensitive work we can use models that don’t retain or train on your data, or run models inside your environment.
What happens to our access at the end?
We hand over all credentials and revoke our own access at handover, unless you’ve chosen a care plan.