Security & Compliance
How we protect your data and keep matters isolated
Security & Compliance
Last updated: August 12, 2026
Company: Percheron AI LLC ("Percheron," "we," "us")
Mailing address: 3400 Cottage Way, Ste G2, Sacramento, CA 95825
Security contact: security@percheronai.com
Percheron handles privileged client material, so we treat confidentiality as a product requirement rather than a configuration option. This page describes the controls we have in place today. It is a summary of our engineering practices, not a contractual commitment — where it touches how we handle personal information, the Privacy Policy governs.
1) Encryption
In transit. All traffic to the application and API is served over HTTPS. Plain HTTP requests are permanently redirected to HTTPS, and our load balancers negotiate TLS 1.2 or higher.
At rest. Uploaded documents are stored in Amazon S3 with server-side encryption (AES-256) applied by default to every object. Our primary database uses encrypted storage backed by AWS-managed keys. Database instances are not publicly reachable — they sit in private subnets and accept connections only from our application security group.
2) Tenant isolation
Every organization's data is scoped to that organization. Application database access is routed through a tenant-scoped session that binds the caller's organization to the connection, and requests carry the organization identifier as a security boundary enforced on each query.
Our vector search index uses native multi-tenancy: each matter's embeddings live in a separate tenant shard, and every search is issued against a single tenant, so retrieval cannot cross a matter boundary.
3) Access & authentication
- Authentication is handled by a dedicated identity provider; we do not store or manage your password ourselves.
- API requests are authenticated with signed, short-lived tokens verified on every call.
- You can review your active sessions and revoke any device you don't recognize from Settings → Security.
- Access to a matter is limited to members of the organization that owns it.
4) Infrastructure & availability
The Services run on Amazon Web Services in the United States. Infrastructure is defined as code and changes are applied through version control and code review rather than by hand.
Our production database is backed up daily with point-in-time recovery over a rolling retention window, and is protected against accidental deletion.
5) Subprocessors & AI processing
We use a small number of vendors to operate the Services, including cloud infrastructure, vector search, and large language model providers. Document content is sent to our LLM provider to power analysis, chat, and drafting.
The current subprocessor list, what each vendor processes, and applicable retention notes are maintained in the Privacy Policy. We keep that list canonical so there is a single place to check.
6) Data retention & deletion
Retention periods for documents, chat history, and account records — and how to request deletion — are described in section 5 of the Privacy Policy.
7) Reporting a vulnerability
If you believe you have found a security issue, please report it to us before disclosing it publicly. We will acknowledge your report and keep you updated as we investigate. We will not pursue legal action against researchers who report in good faith and avoid privacy violations, data destruction, or service degradation.