0 to 1 Product Design
Janie AI Scheduling Tool
Scroll ↓
Role
Product Designer
Team
Heather Sharp (UX), Preston Tyler (PM),
Andrew Krueger (EM)
The Problem
Healthcare offices spend hours manually coordinating patient appointments, follow-ups, and cross-department availability. Scheduling errors, missed notifications, and fragmented communication created operational inefficiencies and increased cognitive load for staff already under pressure. We needed to build a product that enabled users to manage and evaluate the use of an AI scheduling tool.
Design Challenge
How might we design an AI-assisted scheduling experience that feels intuitive, trustworthy, and transparent—empowering healthcare front office workers rather than replacing them?
Research
Methods: Conducted stakeholder interviews, contextual inquiries with front-desk staff, and call flow analysis to understand how scheduling happens during live patient calls.
Findings:
Flexibility and control: Users wanted to easily adjust preferences—such as provider availability, appointment types, or scheduling rules—without needing technical help.
Dynamic data updates: Staff needed a simple way to update the information the AI relies on (e.g., hours, appointment lengths, new providers) to avoid incorrect bookings.
EHR integration transparency: Users were concerned about data accuracy and wanted confirmation that updates were correctly reflected in the EHR.
Trust through visibility: Early prototypes showed that trust increased when users could see a clear audit trail of the AI’s actions—who called, what was booked, and how information was used.
These insights shaped the product’s guiding principles: transparency, editability, and effortless control.
Design Process
The solution to this problem turned out to be quite simple. I created a table that included all of the information that I learned users needed about stalled screenings. I then had several conversations with my dev team and PM to learn about the technical feasibility of the design. It took several conversations to get the whole team on board, but everyone was swayed when they learned we could drastically reduce customer support tickets with this simple table.
This design was tested in 10+ user tests, and our customers were thrilled at the concept, to say the least.
Final Product
This project has been a masterclass in designing for transparency in AI. Building trust doesn’t come from flashy visuals—it comes from giving users visibility and control, while making sure they don’t feel replaced.
I also learned how valuable early design systems can be in AI-driven products; establishing consistent interaction patterns early accelerated development and improved team alignment.
As the tool continues to evolve, I’m excited to see how these foundations help users manage, trust, and fine-tune their AI assistant with confidence.