Sub-processors
Last updated: June 21, 2026
Ken from Finance is operated by Roulette Technologies Pvt Ltd. To provide the Service we engage a small set of third-party sub-processors that may process customer data on our behalf. We require each sub-processor to be bound by data-protection obligations consistent with our own. Data is encrypted in transit (TLS) and at rest, and access within your account is governed by role-based permissions and row-level security.
| Sub-processor | Purpose | Data processed | Location |
|---|---|---|---|
| Supabase | Primary database, authentication, and file storage | Account data, invoice/contract files and extracted data, Slack identifiers | United States (AWS); EU / India regions available for enterprise |
| Vercel | Application hosting and serverless compute | Request data in transit (no primary data store) | United States / global edge |
| Google (Gemini API) | AI extraction of invoice and receipt data | Invoice/receipt documents and extracted text (API tier — not used for model training) | United States |
| OpenAI | AI processing | Document/text content submitted for inference (API tier — not used for model training) | United States |
| Slack | Messaging platform integration (core product surface) | Workspace and user identifiers, messages and files in conversations Ken is added to | United States |
| Twilio SendGrid | Inbound invoice email ingestion and notification email | Email content and attachments sent to Ken, recipient email addresses | United States |
| PostHog | Product analytics | Usage events and pseudonymous identifiers (no invoice content) | United States / EU |
| Cashfree | Subscription payments and India GSTIN verification | Billing details; entity registration numbers submitted for verification | India |
| Stytch | Enterprise single sign-on and identity (SAML / OIDC) | Authentication identifiers and email for enterprise SSO users | United States |
AI processing
Invoice and receipt content is sent to our AI sub-processors (Google Gemini and OpenAI) only to process a customer's own request. We use the providers' API tiers, under which content submitted via the API is not used to train or improve their models, and is not shared between customers.
Data residency
Production runs in a single default region. Enterprise customers can elect data residency with hosting in the United States, the European Union, or India, selected at onboarding.
Changes and notice
We will post additions or changes to this list on this page and update the date above. For material changes, we will notify account administrators by email or in-product notice before the new sub-processor begins processing customer data. Questions or objections can be sent to [email protected].