How we handle your data

The short version

GRADE processes two kinds of customer content: published product documentation and inbound inquiry text. Each customer's data is isolated to that customer, is never used to serve or train for anyone else without written consent, and is returned or deleted within 30 days of exit. Every AI-drafted output requires explicit human approval before it goes anywhere.

Architecture choices that reduce risk

• Tenant write-isolation. A customer's data is isolated on writes as well as reads. GRADE support staff who view a customer's workspace do so read-only — an attempt to create or modify another tenant's data is refused at the application layer, not merely by policy.
• Deterministic screening. Hard requirements (regulatory limits, specifications) are enforced as database filters — not model judgment. Every elimination carries an auditable reason.
• Verbatim provenance. Extracted values must trace to an exact source sentence in the customer's own documents; values without a source are rejected, not stored.
• Human-in-the-loop by design. The system drafts; a named representative reviews, edits, and approves. Nothing sends automatically.
• Release gating. Behavior changes only through versioned releases that pass an automated evaluation suite, including a zero-tolerance check for constraint violations and a standing set of prompt-injection tests.

Operational controls

• Encrypted transport (TLS) on all endpoints; access-controlled application endpoints; rate limiting; structured audit logging that excludes credentials and secrets.
• Encryption at rest. Stored data is encrypted at rest at the storage level by our managed database host; database backups are additionally AES-256 encrypted before they leave the environment, with the decryption key held separately.
• Least-privilege, scoped credentials for infrastructure; secrets kept out of source control; routine dependency vulnerability scanning.
• Subprocessors: model inference via Anthropic under contractual confidentiality; hosting on established cloud infrastructure (Railway, Cloudflare).

Reporting a vulnerability

Email [email protected] with details. We acknowledge promptly, investigate in good faith, and will not pursue action against good-faith research.

Service status

Current operational status, uptime, and any incidents are published on our live System status page.