Industries · MaintenanceOS for property managers
One way of taking repair reports, across every building you manage
Managing property means managing reports from people you rarely meet: residents, leaseholders, tenants of leaseholders, concierge staff, neighbours, the occasional passer-by who spotted water coming through a ceiling. Each building has its own contractors, its own access arrangements and its own history. The risk isn't that you don't care. It's that the way a report is taken depends on who happened to take it. MaintenanceOS gives you one consistent front door for repair reports. The AI takes the details. Fixed rules decide what happens next. A person still clears anything flagged as dangerous.
What makes repair intake hard across a portfolio?
- Many buildings, many callers. The same leak can be reported by three flats in three different ways. Without structure, that becomes three jobs or none.
- Communal areas blur responsibility. A report about a stairwell light, a door entry system or a roof leak needs to land with the right person for that building.
- Urgency drifts. One team member escalates everything, another escalates nothing. Neither is a system.
- Records live in too many places. When a resident, a client or an auditor asks what was reported and when, the answer shouldn't depend on searching inboxes.
How does MaintenanceOS handle a report?
Structured from the first word. The assistant collects the caller's details and turns what they report into a structured repair request, so the job starts with something a contractor can act on.
Rules own the decisions that matter. Safety, the progress of a job and approvals are decided by fixed rules on our server. The AI can't override them, whichever building the call is about.
Dangerous reports stay with a person. Reports our rules flag as safety-critical can enter escalation and human review. They're latched until an authorised person clears them. Our safety rules page explains how this works, and why the list of report types isn't published yet.
UK guidance in an emergency. Where there's an immediate danger to life, the assistant's guidance is 999, never 911.
A record you can stand behind. Consequential actions leave audit evidence alongside the job, and the system only reports something as done when there's evidence it succeeded.
What will it never promise a resident?
The assistant must not invent a booking or appointment time, a contractor dispatch, a refund, compensation, a waived fee, account information or a safety reassurance. If nothing has been arranged, it won't say it has. Not a live always-on phone.
That matters more in block management than almost anywhere. A promise made to one resident is repeated to the whole building by evening. MaintenanceOS also has no authority to spend: quotes and invoices go through checks, but there's no payment, refund or compensation execution in the system. (No client-money authority).
If a caller asks for a person, they can ask to be transferred (live transfer handling is still under testing), and a technical fault is built to fail safe rather than pass the call on unplanned.
How are records organised?
MaintenanceOS is built for many organisations on one system. Your organisation's properties, residents, contractors, work orders, jobs, communications, exports, action evidence and call records sit in your own space, kept apart from other organisations' by isolation controls in the code that are covered by tests. Requests from the phone provider are signature-checked.
There's also an internal supervisor dashboard in the product, which we use to show sample work in a demo. We don't claim a customer portal or client logins today, and we won't until they exist.
What about building safety and housing rules?
Your duties depend on the buildings you manage, who owns them and who lives in them, and they stay with you and your clients whatever software you use. We don't offer compliance promises.
What we do publish is guidance with official sources. Each one is dated and says plainly that it isn't legal advice.
What have we tested so far?
At our 30 September 2026 product checkpoint, all 196 targeted claims-and-safety tests passed, covering explicit transfer, the "never invent" rules, organisation isolation, job lifecycle, decision rules, invoice rules and state-machine controls. Our continuous-integration and security checks have historically passed on the application code too. These are targeted, automated tests run against the product code in a controlled test set-up, not on a live service. They show specific rules behaved as designed in testing, not that every rule will hold in every real situation, and voice intake is still under testing.
Where things stand
We're pre-launch. Voice intake is still under testing, and we're taking early-access requests from property managers. In a demo we'll take a sample report from a block through intake, escalation and clearance, using sample data, and show you what's ready and what isn't.
Frequently asked questions
Can it handle reports for lots of different buildings?
Your properties, residents, contractors and work orders all sit in your organisation's space in MaintenanceOS, and each report becomes a structured repair request. In a demo we'll show how a portfolio would be set up.
Who decides whether a report is urgent?
Fixed rules on our server, not the AI. Reports our rules flag as safety-critical can enter escalation and human review and are latched until an authorised person clears them.
Can the AI approve spending or promise a contractor?
No. It must not invent bookings or dispatches, and there's no payment, refund or compensation authority in the system.
Do clients get a login?
Not today. A customer portal isn't available.
Is it live yet?
Not yet. We're pre-launch and voice intake is still under testing.