Security & compliance

Built safe — at the database layer.

Your buyer data never leaves your tenant. Every query is scoped by tenant and role, and every sensitive action (KYC, payments, bookings) is written to an audit log.

Tenant isolation
TENANT A
Leads
Deals
KYC
Docs
TENANT B
Leads
Deals
KYC
Docs
TENANT C
Leads
Deals
KYC
Docs
Every database query carries the caller's tenant. Cross-tenant reads are structurally impossible.
Three layers of defence

Isolation, guardrails, audits.

Tenant isolation

Every table carries a tenant column. The API rejects any list query without tenant scoping. Cross-tenant reads require an explicit override that only platform admins can issue.

AI guardrails

Every AI prompt runs through a safety layer before reaching the model. You define forbidden topics per tenant — data-exfiltration requests, role-override prompts, legal advice — and set the fallback message.

Audit trail

KYC approvals, deal stage changes, booking confirmations, announcement posts — all logged with who did what, when, and why. Exportable for regulator review.

Role-based access

Four roles. Predictable visibility.

OWNER

Everything in the tenant

ADMIN

Everything except billing

AGENT

Only leads + deals they own

VIEWER

Read-only, scoped like AGENT

UAE data residency

Deploy to your preferred region. Buyer data stays where your regulators require.

PDPL & RERA aware

Audit logs on sensitive reviews. Export-ready for regulatory requests.

No training on your data

Your data is never used to train foundation models. Deletion on request.

Run on the safest AI stack for GCC real estate.

Multi-tenant by default. Role-gated by design. Auditable end-to-end.