.// SECURITY & TRUST

Enterprise security is not a feature. It's the foundation.

Vera was built for regulated industries from day one. Every table, every agent, every action is governed by row-level security, full audit trails, and zero data retention for model training.

SOC 2 Type II GDPR HIPAA ISO 27001

.// COMPLIANCE CERTIFICATIONS

Certified for the most demanding environments.

Vera maintains active compliance certifications across the frameworks that matter most to enterprise security, legal, and compliance teams.

SOC

SOC 2 Type II

Independent third-party audit of security controls, availability, and confidentiality. Continuous monitoring, not point-in-time.

GDPR

GDPR

Full compliance with EU data protection regulations. Data processing agreements, right to erasure, and consent management built in.

HIPAA

HIPAA

Technical safeguards for protected health information. BAA available for healthcare customers. Encryption at rest and in transit.

ISO

ISO 27001

International standard for information security management. Comprehensive risk assessment and treatment methodology.

.// ROW-LEVEL SECURITY & RBAC

920+ tables. 1,607 RLS policies. Zero shortcuts.

Every table in Vera's database is protected by row-level security policies that enforce tenant isolation, role-based access, and agent-scoped permissions. This is not a bolt-on — it's the architecture.

.// ROW-LEVEL SECURITY

Tenant isolation at the database layer

Every query — from human users, API calls, and AI agents — passes through RLS policies that enforce tenant boundaries. A finance agent cannot see HR data, even if it queries the same database. 1,607 policies across 920+ tables ensure complete isolation.

.// ROLE-BASED ACCESS CONTROL

Granular permissions for every role

RBAC policies govern what each user, team, and agent can see and do. Permissions cascade from organization to department to individual — with inheritance rules, override capabilities, and real-time enforcement.

.// AGENT-SCOPED PERMISSIONS

AI agents with least-privilege access

Each of the 41 AI agents operates with a dedicated service role that grants only the minimum permissions needed for its function. The Invoice Matching Agent cannot access customer support data, and vice versa.

.// POLICY ENFORCEMENT

Real-time, not retroactive

Security policies are enforced at query time, not after the fact. If a policy changes, the next query immediately reflects the new rules. No cache invalidation delays, no stale permissions.

Security by the numbers.

Enterprise-grade security infrastructure that scales with your organization.

0+
Database tables with RLS
0
Row-level security policies
0
Agent-scoped service roles
0
Data used for model training

.// DATA RESIDENCY

Your data stays where you need it.

Vera supports configurable data residency for enterprise customers. Choose where your data lives — and prove it to your regulators.

Region Selection

Deploy in US, EU, or APAC regions. Enterprise customers can specify exact cloud regions for data storage and processing.

Data Sovereignty

Meet GDPR data localization requirements with EU-only data processing. No cross-border transfers without explicit configuration.

Compliance Reporting

Generate data residency attestation reports for auditors. Prove where every byte of data is stored, processed, and transmitted.

.// AUDIT TRAILS

Every action. Every agent. Every decision. Logged.

Vera's audit trail system captures every action taken by humans and AI agents — with timestamps, actor identity, data accessed, reasoning chains, and outcomes. Immutable, searchable, and exportable.

.// AGENT REASONING LOGS

See why agents made each decision

Every agent decision includes a reasoning chain — the data it accessed, the logic it applied, and the confidence score. Explainable AI is not optional.

.// IMMUTABLE LOGS

Tamper-proof audit records

Audit logs are append-only and cryptographically signed. No one — not even administrators — can modify or delete historical records.

.// MODEL SECURITY

Your data never trains our models.

Vera uses open-weight models with self-hosted inference. Your enterprise data is never sent to third-party model providers, never used for model training, and never retained beyond the execution window.

Open-Weight Models

Vera runs on open-weight foundation models that can be inspected, audited, and self-hosted. No black-box vendor lock-in.

Self-Hosted Inference

Enterprise customers can run model inference in their own infrastructure. Data never leaves your network for AI processing.

Zero Data Retention

Agent execution contexts are ephemeral. Prompts and completions are purged after execution. Only audit logs persist.

.// ENCRYPTION & INFRASTRUCTURE

Defense in depth at every layer.

From network perimeter to database field — Vera applies defense-in-depth security with encryption, isolation, and continuous monitoring.

TLS 1.3

All data in transit encrypted with TLS 1.3. No fallback to older protocols.

AES-256

All data at rest encrypted with AES-256. Customer-managed encryption keys available.

Network Isolation

Tenant workloads run in isolated network segments. No shared compute or storage.

Continuous Scanning

Automated vulnerability scanning, dependency auditing, and penetration testing on a rolling basis.