1. Hosting and architecture
The platform runs on managed cloud infrastructure operated by tier-one providers with recognised certifications, including ISO 27001 and SOC 2 attestation at the infrastructure layer. Application code runs at the edge; data lives in a managed PostgreSQL service in an access-controlled region. We operate no physical servers of our own.
2. Encryption
- In transit: TLS 1.2 or higher on every connection, with HTTPS enforced and modern cipher suites only.
- At rest: AES-256 encryption for database storage, file storage and backups.
- Secrets: API keys and provider credentials are held in an encrypted secret store, injected only into server-side execution, never shipped to the browser and never written to logs.
3. Access control
- Row-level security is enabled on every application table; records are scoped to their owner and staff access is granted through a dedicated role table checked by a security-definer function.
- Roles are Trainee, Trainer, Manager and Super Admin. Permissions inherit downward only — no account can escalate itself upward.
- Privileged service credentials are used only inside server handlers, after the caller has been verified, and are never used to decide whether a caller is an administrator.
- Staff access follows least privilege and is reviewed when a role changes. Administrator accounts are limited to named individuals.
- Passwords are stored only as salted hashes by our authentication provider. Federated sign-in with Google is supported. Anonymous sign-up is disabled and email confirmation is required.
4. Payments
Card and crypto details are captured on the payment provider's own hosted pages. We never see, transmit or store card numbers, so the platform stays outside PCI DSS storage scope. Provider callbacks are authenticated by signature verification with timing-safe comparison before any data is written, and payments default to test mode until a live switch is explicitly confirmed by a Super Admin.
5. Backups and resilience
- Automated daily database backups with point-in-time recovery on the managed database service.
- Backups are encrypted, access-restricted, and retained on a rolling window.
- Restore procedures are documented and tested; recovery objectives are shared with corporate clients on request.
- Learner progress is also held locally on the learner's device and can be exported as a portable backup file.
6. Vulnerability management and monitoring
- Dependency scanning on the code base, with prompt patching of high-severity findings.
- Automated database and access-policy linting to catch misconfigured permissions before release.
- Runtime error capture and request logging, reviewed for anomalies.
- Input validation with schema checks on every server endpoint; public endpoints verify the caller inside the handler rather than relying on the URL.
- Security findings are triaged by severity: critical issues are remediated as a priority release, others in the normal release cycle.
7. People and process
Everyone with access to production data is bound by confidentiality terms and works on company-managed accounts with multi-factor authentication. Changes reach production through reviewed releases. Suspected incidents follow a documented response process: contain, assess, notify affected clients within seventy-two hours, remediate, and record lessons learned.
8. Responsible disclosure
Report a suspected vulnerability to petrazim.solutions@gmail.com with enough detail to reproduce it. Please do not access other users' data, degrade the service, or disclose the issue publicly before we have fixed it. We acknowledge reports within two business days.