Healthcare
WhatsApp-Native Agentic AI for a Surgical Practice
We built a WhatsApp-native AI assistant for a surgical practice. It now handles routine patient conversations, scheduling, and estimate generation with human escalation when needed.
Industry
Healthcare (Specialty Surgical Practice)
Client
A surgical specialty practice under NDA, engaged via a strategic AI partner
Engagement
End-to-end agentic AI build deployed on WhatsApp with iterative accuracy improvement
Outcome
Live with the first client deployment in production, handling patient inquiries indistinguishably from human staff
The Challenge
A surgical specialty practice was losing time to high-volume, repetitive patient communication: pre- and post-surgery questions, appointment scheduling across multiple clinic locations, and personalized cost estimates that staff had to assemble manually for each inquiry.
The doctor needed to scale patient responsiveness without scaling headcount, and patients needed to keep using WhatsApp, the channel they were already on.
What We Built
- Designed and built an agentic AI system that operates inside the doctor's WhatsApp account and handles patient interactions directly.
- Covered three core agent functions: routine medical Q&A, appointment scheduling across two clinic locations, and personalized surgery cost estimation generated as PDF documents.
- Kept the UX purely conversational, with no bot menus or structured forms that would break the natural feel of the interaction.
- Added graceful human escalation with full conversation context preserved when a query moves outside the agent's scope.
- Grounded responses in approved material through retrieval over the practice's medical content and pricing logic.
- Connected live scheduling into Google Calendar across both clinic locations, including conflict handling.
What Changed
- The system is live in production after iterative accuracy improvements made it reliable enough for clinical communication.
- Patient-facing communication scaled without requiring equivalent staff growth.
- The practice gained a WhatsApp-first delivery model that meets patients on the channel they already use.
- The underlying architecture became reusable for adjacent healthcare workflows.
Stack
Next step
Patient or customer communication eating your team's day?
We can map the workflow, tell you whether this pattern fits your operation, and outline what a first delivery slice would look like.
Discuss a similar build