Shengul AI
Sign inBook a meeting

Last updated 25 July 2026

Subprocessors

These are the companies we rely on to run Shengul AI. Each processes personal data on our instructions, under a written contract with confidentiality, security and data-protection terms at least as protective as our own.

On this page

  1. 1. Current subprocessors
  2. 2. Model training
  3. 3. What we require of them
  4. 4. Changes to this list

1. Current subprocessors

ProviderWhat it does for usPersonal data involvedPrimary location
Supabase, Inc.Database, authentication and file storage — the system of record for accounts, campaigns, contacts, research and email threadsAll categories held by the serviceUnited States
Vercel, Inc.Application hosting, serverless execution and content deliveryAccount data, technical and log data, and any data in transit through the applicationUnited States
Apollo.io (ZenLeads, Inc.)Sourcing business contacts matching a client's buyer profile, and revealing and verifying work email addressesBusiness contact details — name, job title, employer, work email, professional profileUnited States
Emailable, Inc.Independent second check that a work email address is deliverable before anything is sentWork email addressesUnited States
Bright Data Ltd.Retrieving publicly available web results used to research a target companyCompany and contact names used as search termsIsrael, with infrastructure in the United States and the European Union
Google LLCGemini large language models, used to draft research summaries, emails, follow-ups and replies; and the Gmail API, where a client has connected a Google mailboxContact details, research dossiers and email content sent as model input; mailbox content for connected Google mailboxesUnited States
Microsoft CorporationMicrosoft Graph mail APIs, where a client has connected an Outlook or Microsoft 365 mailboxMailbox content for connected Microsoft mailboxesUnited States
Upstash, Inc.Scheduling and delivering the background jobs that run the pipeline, including delayed follow-upsRecord identifiers only — no contact details or message contentUnited States

Where a client connects a mailbox using SMTP and IMAP instead of Google or Microsoft, their own mail host processes the messages. That host is chosen by the client, not by us, and is not our subprocessor.

2. Model training

We use the language-model providers above under terms that prohibit them from using our data — including client content and contact data — to train their general-purpose models. We do not train any model on one client's data for the benefit of another, and we do not sell or license client data to anyone.

3. What we require of them

Before we engage a subprocessor we assess its security and privacy practices, and we put in place a written contract that:

  • Limits processing to our documented instructions and the purpose described above.
  • Imposes confidentiality obligations on its personnel.
  • Requires appropriate technical and organisational security measures.
  • Requires it to assist us with data-subject requests and breach notification.
  • Requires deletion or return of the data when the engagement ends.
  • Applies equivalent obligations to any onward subprocessor it uses.
  • Includes an approved transfer mechanism where personal data leaves the EEA, UK or Switzerland.

We remain responsible to our clients for a subprocessor's performance of these obligations.

4. Changes to this list

We will give clients at least 30 days' notice before adding or replacing a subprocessor, by updating this page and emailing the notice addresses on the account. To receive that notice, email support@foundersideai.com and ask to be added to the subprocessor notification list.

A client may object on reasonable data-protection grounds within those 30 days, as described in section 6 of the Data Processing Addendum at /legal/data-processing-addendum. If we cannot resolve the objection, the client may terminate the affected part of the Services and receive a refund of fees prepaid for the unused period.

We may engage a replacement subprocessor immediately, without prior notice, where it is necessary to keep the Services running or to address a security risk. We will update this page and notify clients as soon as we can afterwards.

Contact

Exportpoint LLC1209 Mountain Road Place Northeast, Albuquerque, NM 87110Albuquerque, NM 87110United Statessupport@foundersideai.com(+1)+19292141601
Shengul AI
How it worksPrivacy and securitySign inBook a meeting
Privacy NoticeIf you received an email from usCookie NoticeTerms of ServiceAcceptable Use PolicyData Processing AddendumSubprocessors

2026 Shengul AI. Outbound, handled.