As a Staff Product Designer in ServiceNow's Healthcare & Life Sciences group, I lead design for how AI agents work inside regulated care and research, across two products: Care Team Operations at the bedside, and Clinical Trial Operations for multi-site Phase III trials.
A nurse at the bedside and a trial lead running fourteen sites have very different days. The AI in both products follows the same three rules.
The Voice Agent captures, classifies and dispatches a broken pump. Otto flags a trial site that's likely to slip. In both, a person decides what happens next.
Nurses flag issues and see status inside the EHR. Trial leads see every site, team and blocker in one workspace. Nobody learns a new tool to get the AI's help.
Live status, an explanation for every flag, the health of every connected system, and an audit trail a team could defend in an inspection. In regulated work, an AI that can't show its reasoning doesn't get used.
The patterns I build here reach past these two products.
ServiceNow is building an AI framework into its Horizon design system. Developers across products and business units adopt it alongside the Lit library to build faster. The workflows I'm designing feed into it, so what works at the bedside and in trial operations can shape platform-level workflows.
Every month I run a working session with designers from other teams. We share component work and workshop agentic experiences that cross product lines, so the places where our products connect work as one system.
Both are password-protected because the work is confidential. Email me at hello@araxiemiller.com for access.
Resolving the floor's breakdowns from the bedside: voice capture, smart routing, and live status inside the EHR.
Activating multi-site Phase III trials: unified orchestration and AI-driven risk detection for clinical operations leaders.