Trust Center
Sub-processors
The full list of third parties that process data to deliver the service. The same sub-processors and the same wording as Annex 2 of the data processing agreement — both tables read one register, so they cannot drift apart. The transfer mechanism is on a page of its own, so it is visible without reading the contract.
The sub-processor list
| Sub-processor | Purpose | Data | Location | Engaged | Where this is written down |
|---|---|---|---|---|---|
| Hetzner Online GmbH | Infrastructure, compute, databases, storage and backups for the website and the platform | Customer content, account and application data, files, logs | EEA Awaiting confirmation: the specific region and availability zone | Always | Security Summary → Infrastructure and data residency |
| Cloudflare, Inc. | DNS, content delivery, WAF, DDoS mitigation and bot protection Awaiting confirmation: which of these are enabled in front of the site today, since only Turnstile is proven | IP address, request headers, request and security metadata | EU / global edge, depending on configuration Awaiting confirmation: the edge configuration in force | Always | Privacy Policy → International transfers |
| Scaleway SAS | AI inference using agreed models on Scaleway infrastructure | Prompts, selected context, files where applicable, generated output | France / EEA Awaiting confirmation: the specific region | AI features | DPA → Annex 2 |
The “Awaiting confirmation” marker means exactly what it says: this is the owner’s wording, checked against neither a contract nor a deployment. We show it as an open question rather than as a fact — and we name which question is open. The company names are owner-confirmed; regions, zones and what is switched on right now are not.
Every sub-processor is bound by written data protection terms at least as protective as our data processing agreement. We remain fully liable to you for their acts and omissions as if they were our own.
For 1 of the 3, processing leaves the EEA, and “infrastructure in the EEA” does not cover that: the mechanism for each transfer is on a page of its own.
What happens when the list changes
Article 28(2) GDPR requires prior notice before a new sub-processor is engaged. A published list creates a second obligation — keeping it current. Both are written into the data processing agreement, not only onto this page.
- before processing starts
Warnings
Before a new sub-processor starts processing your personal data, we send you notice — before, not after.
- within the notice period
Your objection
You may object on reasonable data-protection grounds. We will work with you in good faith to find an alternative.
- if none is found
Exit without penalty
If no alternative is found within the same period from your objection, you may terminate the affected part of the service without penalty and receive a pro-rata refund of prepaid fees for it.
The verbatim wording is in the “Sub-processors” section of the data processing agreement: open the section.
Who is not on the list, and why
The entries below are not on the list. For each of them the reason is named.
-
Workflow automation
The system that receives and routes enquiries from the forms is open-source software we deployed and operate ourselves, on the infrastructure of our hosting provider inside the EEA. It is us processing the data rather than a third party we hand it to, which is why it is not on the sub-processor list — while the provider whose infrastructure it runs on is. -
An AI provider of your own
Where you connect an AI provider through your own account — Companion works this way — the nature of the processing relationship depends on your configuration and your agreement with that provider. Sintora does not use an account of its own for that inference route and does not engage that provider — and this list is of the parties we do engage. -
A transactional email provider
We do not use one yet. As soon as we do, it joins the list under the same notice period as any other. -
The developers of the models our inference provider runs
A sub-processor is a party your data reaches. Where our inference provider executes a model’s weights on its own infrastructure, the model’s developer receives none of your prompts, none of the context and none of the output — so it is not our sub-processor and it is not on the list. The model families themselves we name separately, on the “AI and your data” page: “whose model” is worth answering, but the answer is not a line in the register.
Want to hear about changes first
Write to us and we will notify the address you give us before the list changes. There is no signup widget here: a subscription with confirmation and unsubscribe needs a store of subscribers, which this site does not have.
Changes to the documents will be recorded in the log with a date: the change log.