Security
How RecruitSecure AI protects your data at every layer.
Data Architecture
RecruitSecure AI uses a multi-tenant architecture with strict data isolation. Each organization's data is separated at the database level using application-level organizationId filtering on every query, ensuring that one tenant can never access another's records.
All data is encrypted at rest using AES-256-GCM and in transit using TLS 1.3. Database is hosted on Supabase with automated daily backups and point-in-time recovery.
AI Privacy
No external AI APIs. The embedding model used for semantic search runs entirely within your deployment — whether on Vercel serverless functions or your own servers. No data is sent to OpenAI, Google, or any third-party AI provider.
The model generates vector embeddings from CV text locally, and those embeddings are stored in your tenant's isolated database partition. No candidate data leaves your infrastructure for AI processing.
GDPR Compliance
- Data Export: Users can request a full export of their data at any time through the dashboard or by contacting support.
- Right to Deletion: All candidate data, including CVs, embeddings, and metadata, can be permanently deleted upon request. Deletion is propagated to backups within 30 days.
- Consent Management: The platform provides tools for managing candidate consent, including recording consent basis, purpose, and withdrawal.
- Data Processing Agreement: A DPA is available for all customers. Enterprise customers receive a customized DPA as part of their agreement.
EU AI Act Ready
RecruitSecure AI is designed with the EU AI Act requirements in mind. As an AI system used in employment and recruitment, we implement the following safeguards:
- Audit Trail: Every AI-driven search and ranking decision is logged with timestamps, query parameters, and scoring details for full traceability.
- Human Oversight: AI results are presented as recommendations. All hiring decisions remain with human recruiters. The system does not make autonomous decisions.
- Transparency: Candidates can be informed about the use of AI in the recruitment process. Similarity scores and ranking factors are visible to recruiters.
Infrastructure
RecruitSecure AI is built on a modern, security-focused stack:
- PostgreSQL + pgvector on Supabase — EU region available for data residency compliance.
- Transformers.js with all-MiniLM-L6-v2 — 384-dimensional embeddings generated locally. No external AI calls.
- AES-256-GCM encryption — all personally identifiable information is encrypted at rest.
- Auth.js v5 with JWT strategy — credentials, Google, and Microsoft authentication supported.
- Stripe — PCI-DSS compliant payment processing.
- Sentry — error monitoring with sensitive data scrubbing.
- PostHog — product analytics hosted on EU servers with a privacy-first configuration.
- Upstash Redis — serverless rate limiting to protect against abuse.
- Resend — transactional email delivery.
Data Residency
For EU customers, data is stored in Supabase EU region data centers to comply with data residency requirements. Enterprise customers can request specific data residency configurations as part of their deployment.
Responsible Disclosure
If you discover a security vulnerability, please report it responsibly by emailing contact@recruitsecureai.com. We take all reports seriously and will respond within 48 hours. We ask that you do not publicly disclose the vulnerability until we have had a chance to address it.