Security
Confidence to put AI
in a regulated workflow.
Revo is engineered for regulated engineering environments. AI proposes, humans approve, and the parts that must be reproducible — like wiring — are handled by fixed logic, not re-imagined at view time.
Multi-tenant architecture
Every customer organization runs in a cryptographically isolated tenant with dedicated data boundaries.
Human-in-the-loop safety
The design cannot silently advance past a step that still needs human sign-off — an auditable trail follows every AI decision.
Deterministic wiring output
Wiring is generated by fixed logic from a one-time AI netlist. Same inputs, same output, byte for byte.
End-to-end encrypted
TLS 1.3 in transit, AES-256 at rest — for every BOM, drawing, and schematic.
SSO & role-based access
SAML/OIDC single sign-on, SCIM provisioning, and granular Roles & Permissions across the workspace.
Immutable audit log
Every AI run, every override, every user action — permanently traceable at the platform admin layer.
Metered AI credits
AI usage runs on a credit system so cost stays predictable and controllable per tenant.
Design snapshots
Placement approvals are recorded as locked read-only snapshots — a clear 'this was approved' record.
Principles
Data handling.
Tenant isolation by design. Customer organizations are separated at the storage and processing layer. Nothing is shared across tenants.
AI proposes, humans approve. No AI step advances the design without explicit user sign-off. Approvals become locked snapshots for audit.
Deterministic where reproducibility matters. Wiring output is byte-for-byte reproducible. A failure or slowdown in AI can never corrupt the schematic.
Metered, predictable AI usage. Tenants control AI cost via a credit system with per-workspace and per-member limits.