Security

Enterprise-Grade Protection

Resident-facing AI carries your brand and your residents' data. Here's how we keep both safe.

SOC 2Type II
PIPEDA
CASL

Encrypted end to end

Every request runs over TLS and everything at rest is encrypted with AES-256, on infrastructure we monitor around the clock.

Least-privilege access

Access is scoped to the buildings a person actually works on, so a site manager never sees another property's residents. Every entry is logged and reviewable.

No model training

Your conversations, documents, and resident data are never used to train a model for anyone but you. Nothing you upload improves someone else's agent.

You own your data

All prospect and conversational data belongs to you. Export it in full whenever you want, and it is never sold, brokered, or shared with a third party.

Evaluated on every message

A seven-step evaluation pipeline runs on every message before it reaches a resident, and a regression suite runs on every code change so agent behaviour doesn't drift.

Humans in the loop

When confidence drops or a conversation turns sensitive, the agent stops and escalates to your team with the full thread attached, rather than guessing.