🔒
Anonymised before it leaves
Sensitive records are anonymised before a model sees them, so what goes out is the question, not your customer.
01Benefits
Why teams love it
What you get out of the box. No premium tier.
🌍
Data residency
Frankfurt, US East and Melbourne today, and a new region when a customer needs one.
🧾
Every action on the record
Whatever AI writes lands in the same immutable audit trail as every other change.
02Mechanics
How it works
From signal to outcome in four explicit steps, with no hidden hand-offs.
Ask inside the platform
The Onboarding Agent sits where your team already works and answers from your own data.
Or connect your own tool
Claude, ChatGPT or Cursor reach the platform through the standard MCP endpoint.
Anonymised on the way out
Sensitive records are anonymised before a model sees them, and processing stays in your own jurisdiction.
Everything written is logged
Whatever AI changes lands in the immutable audit trail, like every other write.
03Capabilities
What's inside
Every headline backed by a concrete workflow. Screenshots show what your team will actually use.
Capability 01
The Onboarding Agent, inside the platform
Ask about your own data where you already work: which onboardings are blocked, how many are done, what a customer is waiting on. It also drafts and sharpens the text you would otherwise write yourself.
Capability 02
Or your own AI tool
Claude, ChatGPT and Cursor query and act on the platform through the standard MCP endpoint. Ask about an onboarding, create tasks, pull status, without leaving the tool your team already uses.
Want to try it against your own data?
20-minute call. We connect your AI tool to a test tenant and you ask it whatever you would ask a colleague.
Capability 03
Made for sensitive data
Sensitive records are anonymised before a model sees them, and processing stays in your own jurisdiction: Frankfurt, US East or Melbourne. If your customers need a region we do not run yet, adding one is a matter of setup, not of architecture.
04Use cases
Situations we keep seeing
Anonymised from our customer base. See whether one of them is yours.
Legal-tech compliance lead
Scenario
AI sensitivity is high. Every AI action has to be reviewable for client trust audits.
Outcome
Anonymised input, inference in the same jurisdiction, every action in the audit trail. AI became a question of settings rather than principle.
Mid-market CS team lead
Scenario
The team lives in ChatGPT and wanted their onboarding data available there too.
Outcome
MCP access lets them query state and create tasks from the chat they already have open.
05What we replace
What you often stop needing
Three habits that quietly cost more than they look like.
Pasting into off-platform ChatGPTNo context, no record, and customer details in a chat nobody can audit afterwards.
Bolt-on AI assistantsGeneric tools without your operational data. They cannot see a flow or a blocked onboarding.
Asking the colleague who knowsEvery question routed to the one person who knows where to look. Fine until they are on holiday.
06Across the platform
Where AI already does the work
AI is not a separate room in the platform. These capabilities use it inside their own job, with no extra tier and no separate product.
Frequently asked
No. The Onboarding Agent is part of the platform. Connecting your own tool through MCP is an option for teams that would rather stay in Claude, ChatGPT or Cursor.
That is what it is built for. Sensitive records are anonymised before a model sees them, processing stays in the region you choose, and every AI action is recorded in the audit trail your auditor already reads.
In one of three regions today: Frankfurt, Northern Virginia or Melbourne. You pick the one your customers require, and a further region can be set up if they need one. Sensitive records are anonymised before they reach a model.
Anything that speaks MCP. Claude, ChatGPT and Cursor are the ones our customers use today.
Ready to see it on your data?
20-minute call. Show us your source, we show you the mapping. No deck required.