We built this for PetSure.
This isn't a slide deck. It's a working AI agent that handles pet parent questions about claim status and lodging a new claim — pulling live from a mock PetSure data layer. Try it below. The chat reads your real help-centre style content and the workflows match how a PetSure agent should actually behave.
The voice version of this agent runs on the same workflows and KB. In production it answers the PetSure main line 24/7 — handles claim status checks, takes a new claim by phone, reads claim numbers and dollar amounts digit-by-digit, and reassures distressed pet parents while their pet is at the vet.
How we built this
Honest version — no smoke. Here's exactly what's behind this chat:
Scraped your help centre
Crawled petsure.com.au public claims and FAQ content into the agent's knowledge base. The agent quotes from this when pet parents ask about waiting periods, what's covered and how the claim process works.
Built workflows around your model
Wrote natural-language workflows for claim status, lodging a new claim, and routing. Every workflow quotes claim and pet details back to the parent before any destructive action and never narrates "let me check".
Connected mock PetSure tools
Wired in three mock tools — pet parent profile, claim status, file a new claim — that mirror real APIs. In production these point at your real systems with PII handling we already do for our regulated subscribers.
Went live
Published all workflows to the live environment with guardrails active — knowledge gap handling, no-weak-uncertainty steering, and de-escalation for distressed pet parents. 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 KB, real claims data, and the underwriter integration takes a few weeks. We've done it for regulated insurers across AU, US, and EU.
Discuss with our experts →
Demo built with