Skip to content

Property management · AI automation

Maintenance emergencies triaged in each company's own words

Resident maintenance requests, owner questions and viewing requests all land in one place, and a routine request looks the same as an urgent one.

One engine reads each property company's own emergency words, skips scheduling for emergencies and routes requests to the right person.

This is a concept build. Kestrel Property Management is a fictional business. The code is real: open the demo and try the steps below.

The everyday problem

Resident maintenance requests, owner questions and viewing requests all land in one place, and a routine request looks the same as an urgent one.

Each request is read in the company's own words, and emergencies jump the line.

The hard part

  1. Emergencies that differ by business

    Kestrel lists 'no heat', 'carbon monoxide' and 'cannot lock'. Dune Lake lists 'locked out', 'hurricane' and 'evacuation'.

  2. An emergency offered a calendar

    Emergencies skip slot picking and go to the emergency route.

  3. The wrong person gets the job

    Assignment prefers the right skill, then the right area, and takes whoever has the most room. If nobody fits, it widens to the whole team.

  4. Two viewings for one slot

    The database refuses the second booking. The request is kept.

How it runs

Workflow

How a maintenance request finds the right person.

Sample run
  1. Step 5: Hold the slot. Viewing and service slots cannot collide.
    If this fails: Slot just taken: pick another, request kept.

Tap “On fail” to see what happens when a step breaks.

Edge cases we handled

What the system does in the unusual cases. Where there is a phrase, type it into the demo.
WhenWhat the system doesTry it
A resident writes 'no heat and it's freezing tonight'.Read as an emergency. Scheduling is skipped.no heat and it's freezing tonight
A guest is locked out of a vacation home.Read as an emergency on that company's list.No phrase to try
Everyone with the right skill is full.Held in the office queue, and logged.No phrase to try
Two people book the same slot.The second is asked to pick another. The request is kept.No phrase to try

Guardrails and hand-off

Guardrails

  • Emergency words come from each company's own knowledge file.
  • Emergencies never wait for a calendar slot.
  • The database decides who gets a slot.

When a person takes over

Emergencies go to the emergency route for the office, with the request and its score. Anything the rules cannot place waits in the office queue.

Not in this build yet

We would rather you hear it from us.

  • Fair Housing rules and three-audience routing exist as written assistant instructions, not in the running demo.
  • There is no AppFolio or Buildium connection.
  • Vendor dispatch is logged only. No vendor is contacted.

Try it yourself

  1. Type: no heat and it's freezing tonight

    Expect: Read as an emergency. Scheduling is skipped.

Open the chat on the demo site. The chat button appears once you have scrolled a little, or after about 20 seconds. Please don't type real personal details.

Built with

  • TypeScript
  • Next.js
  • React
  • Not in this concept build. Emergency alerts and vendor dispatch are logged for the office. In your build, we connect the tool you already use.

  • Yes. It reads one knowledge file per business, so your emergency words, service area and team drive it.

More work

How Ridgewell intake engine runs. Steps: Lead arrives, then Read urgency, then Match the service, then Score 0-100, then Pick a crew, then Hold the window, then Office crew board.

  • Concept build
  • Roofing
  • AI automation

Roofing leads that know an emergency when they read one

Everyday: Web leads sit in one inbox and get answered in the order they arrived. In a storm week, the leaking ceiling waits behind the gutter quote.

The hard part: An emergency typed casually

How Rnext service engine runs. Steps: One knowledge file per business, then All read the same file, then Business id added on the server, then Own database per business, then Release automation.

  • Our own product
  • Multi-trade
  • AI automation

One engine, six businesses, and a price that never disagrees with itself

Everyday: The website says one price, the estimator says another and the chat says something else. Every new location means rebuilding the logic.

The hard part: The estimator and the price table disagree

How Everline home search runs. Steps: Question, then Read the budget, then Match the market, then Skip sold listings, then Up to three listings, then A person on request.

  • Concept build
  • Real estate
  • AI services

Home search by plain-language question, with an honest 'nothing fits'

Everyday: Buyers ask 'anything in Austin under 1.2m?' in chat and get a link to a search form.

The hard part: A budget typed loosely

Start a project

Want something like this for your business?

A 30-minute call, then a written plan and a fixed quote. No obligation.

  • A reply from a person within one business day.
  • Fixed price in writing
  • You own everything