Case Study · 03 / 05
AI Scheduler.
An agent that produces ranked appointment options in seconds.
Context
Where they were coming from.
Schedulers at a specialty clinic were spending 30 to 40 minutes per appointment juggling provider availability, room constraints, and patient preference. The job was a calendar Tetris game played by humans.
The judgement calls were small. The constraint-satisfaction was enormous.
What We Built
The agent, specified.
An agent that ingests the patient's preferences, the provider constraints, the room and equipment requirements, and returns three ranked appointment options in under five seconds. The scheduler picks; the agent books.
- Calendar integration with read/write access
- Constraint engine for provider, room, equipment, and patient preference
- Ranked output with rationale (so a human can override with context)
- Audit log of every booking decision
Results
The shape of the curve after launch.
Illustrative metrics pending client sign-off. The shape of the curve matters more than the exact number. These are the moves we aim for.
4 min
Down From 38 Min Per Appointment
+22%
Daily Slot Utilisation
0
Double Bookings Since Launch
Have a workflow like this?
Bring us the workflow that never quite gets done. We'll spec the agent that does it.