We built this for InteliCare.
Not a deck. A working AI agent that handles the questions adult children send to InteliCare support — what an alert means, how to tune thresholds so it pings less, and account questions like pausing the service when mum's in hospital. Try it below. The chat reads from an InteliCare-shaped knowledge base and the workflows match how a thoughtful support specialist should actually answer.
The voice version of this agent runs on the same workflows and KB. In production it answers the InteliCare support line — handles alert questions, walks through threshold tweaks by voice, and reads sensor IDs digit by digit when troubleshooting.
How we built this
Honest version. Here's what's behind this chat:
Wrote InteliCare-shaped knowledge
Five hero articles covering what the sensors detect (and what they don't), threshold tuning, plan pricing, pause policies, and sensor troubleshooting. The agent quotes from these instead of guessing.
Built workflows around your model
Natural-language workflows for alerts/sensors and account/subscription. The alerts workflow reads the family's recent alerts and tunes thresholds inline — never makes a clinical judgement from sensor data alone.
Connected mock InteliCare tools
Three mock tools — family profile, recent alerts, adjust threshold — that mirror real APIs. In production these point at your real systems with the PII handling we already do for regulated subscribers.
Went live
Published all workflows to the live environment with guardrails active — knowledge gap handling, no-clinical-judgement steering, and a steer-away from hedging. This page talks to that live agent.
Want to build the real one?
This took us ~90 minutes with one engineer. Production with your real sensor telemetry, family records, and the care advisor calendar takes a few weeks. We've done it for regulated healthcare and aged care subscribers across AU, US, and EU.
Discuss with our experts →
Demo built with