Security and ownership
Yours, visible, and safe to fail.
Three principles first. The detail is underneath for anyone who wants it.
Three principles
You own it.
Where practical, the source, the hosting and the service accounts sit under your company's control from the start.
The data path is visible.
A written map of where each kind of data goes, including which AI step sees which fields.
Failure has a manual path.
Critical workflows have monitoring, retries where safe, visible failure states and a documented fallback.
How each part is handled
Source, accounts, credentials
Where practical, production infrastructure, source repositories and service accounts are created under the client's control. Handover includes what another competent team needs to operate and maintain the system.
Who sees and changes what
Each person sees and changes what their role needs, and approval limits are set per role. Our access is personal, never shared logins, and is removed when the engagement ends.
Selected fields only
Only steps that need AI send anything, and only the fields that step needs. Fields you mark sensitive are masked first. The data-flow document names the provider and model for each step and its retention terms.
The operating record
Records, attachments, rules and approval limits sit in a database in your account, exchanging data with your ERP, accounting system and email through their APIs.
Who did what, when
Every change records who or what made it, when, and the before and after values. The log cannot be edited from the application.
A named person decides
Prices below margin, purchase orders, quality dispositions, invoice release and any message to a customer on a new basis can be set to wait for a named person, with a delegate if they are away.
Defined per system
Backup and recovery are defined for each production system according to its hosting and importance. The recovery procedure is documented and tested before handover.
What handover includes, where applicable
- Source code and repository access
- Credentials and account inventory
- Deployment and restart runbook
- Dependency and integration inventory
- Documented data flow
- Role and approval configuration
- Manual fallback procedure for critical workflows
- Backup and restore procedure
- Monitoring and failure-handling documentation
- Handover walkthrough
Questions a careful buyer asks
Who owns the source code?
You do. It lives in a repository in your organisation's account from the first day, and the licence is assigned to you in the contract.
Who holds the credentials?
You do. We keep a credentials inventory: every system, what access it has, who holds it, and how to rotate it. Our access is personal and is removed when the engagement ends.
What happens if Bubbletech disappears?
The system runs on accounts under your control where practical, and handover includes what another competent team needs to operate and maintain it.
What happens if an automation fails?
Critical workflows have monitoring, retries where safe, visible failure states and a documented manual fallback. Steps are designed so a retry cannot post an invoice or send a message twice.
Can we operate manually?
Yes. Every workflow has a written manual procedure. If the system is unavailable, work continues and is entered afterwards; the gap shows in the audit trail.
Which AI provider sees which data?
Only steps that need AI send anything, and only the fields that step needs. The data-flow document names the provider and model for each step. We use providers whose API terms do not permit training on your inputs by default.
Are sensitive fields redacted?
Fields you mark as sensitive, such as personal phone numbers, bank details or patient information, are excluded or masked before any AI step.
Are approvals human-controlled?
Consequential actions wait for a named person: prices below margin, purchase orders, quality dispositions, releasing invoices, and any message to a customer on a new basis. You decide where the lines are.