Trust
Security
Last updated:
Voxaris operates production AI infrastructure that handles customer data, voice calls, SMS, and CRM webhooks for service businesses. Security is treated as a product requirement, not an afterthought.
Reporting a vulnerability
Email security@voxaris.io or admin@voxaris.io. Targets: acknowledgment in 2 business days, triage in 5, critical patches in 14. See /.well-known/security.txt.
How we secure data
Encryption. TLS 1.2+ on every surface, 2-year HSTS preload, secrets in environment-scoped secret managers, never in source control.
Database. Row-level security on every multi-tenant table. Cross-tenant reads are not possible.
Voice & SMS. LiveKit Cloud + Twilio Elastic SIP. Outbound voice requires per-lead TCPA voice-consent (FCC AI-voice disclosure, Feb 2024). Outbound SMS handles STOP, HELP, and quiet hours.
Code. Builds gate on CI (build + typecheck + lint). Security-relevant changes get a second-engineer or agent review before merge.
Subprocessors
- Vercel — hosting + edge
- Supabase — Postgres + auth + storage
- Neon — Postgres (audit.voxaris.io)
- Twilio — SMS + voice telephony
- LiveKit Cloud — voice AI media routing
- Google Cloud — Gemini + Solar + Maps
- Anthropic — Claude API
- OpenAI — speech-to-text + select inference
We do not sell or rent customer data to any third party.
Acknowledgments
None yet — be the first.