As a UX and Product Designer, I designed and specified a complex medical SaaS for clinical teams in a team with Marcel Caspi (PRODUX)
2021 – 2023 | Team collaboration with Marcel Caspi, PRODUX
1. Complex responsive system design and developer handoff
2. Feature specification and design for new features, alongside system maintenance
1. Complex responsive system design
Therapy Manager is a SaaS product for medical staff to manage and monitor patient therapy progress.
In the first year of the engagement, the work was primarily UI-intensive: designing and building a complex responsive system across desktop, mobile, and tablet, and preparing complete developer handoff.
The system had to work across device types without losing clinical clarity. Medical staff access it from different contexts, each with different screen constraints and workflow rhythms. Getting the responsive behavior right wasn't a styling decision. It was a structural one.
The Challenge
At Medical SaaS system the interface has to be clear under pressure, consistent across device types, and precise in how it surfaces information. At the same time, it had to integrate into an existing product ecosystem, not replace it.
The challenge was to build a complex, multi-screen system that felt coherent and low-friction for medical staff, across desktop, mobile, and tablet, while setting the technical foundation for future feature work.
The process
The work followed a structured build approach:
- Mapping the system architecture and information hierarchy before designing individual screens
- Designing responsive behavior intentionally: understanding which information needed to be present at each device tier, and what could be de-prioritized without losing clinical utility
- Maintaining consistency across a large system by building from reusable components and clear patterns
- Preparing complete developer handoff documentation so the implementation matched the design intent
All of this was done in close collaboration with Marcel Caspi, senior UX strategist, who led the strategy and specification layer.
The Solution
A complex responsive system designed across three device types: desktop, tablet, and mobile. The system was built for clinical staff navigating patient therapy data in real conditions, not ideal ones.
Each screen was designed to surface the right information at the right level, without overloading the interface.
Developer handoff included complete specifications, component documentation, and interaction notes.
The result
A fully designed, developer-ready responsive system covering the core Therapy Manager product. The handoff enabled the development team to implement with precision, and the foundation built in this phase supported the feature work that followed.
Dosentrx, Therapy Manager medical portal
Complex responsive system design for medical staff. Feature specification from diagnosis to delivery. Cross-system UX mapping between SaaS portal and physical device. A principled approach to feature scoping: understand before designing.
Done in close collaboration with Marcel Caspi, PRODUX.
Feature specification and new feature design
In the second year, the focus shifted from building the system to extending it. The work became more pure UX: diagnosing where new features needed to go, mapping what already existed, and specifying what had to change.
One documented example captures the approach well: mapping what happens in the system when a patient needs a new medication cassette for the ReX device.
As with all work on this engagement, this was done in close collaboration with Marcel Caspi (PRODUX).
The Challenge
A new scenario entered the product: a patient needs a medication cassette replacement for their ReX device. On the surface, it sounded like a small addition. In practice, it touched multiple parts of the system, crossed into the patient journey, and raised the question of whether a physician needed to be involved.
The challenge wasn't designing the solution. It was first understanding what the problem actually was.
The process
The process started with diagnosis, not design:
Collected all relevant information from the product manager
Mapped the user journey of the medical staff working within the system
Mapped where this journey intersects with the patient directly
Identified whether physician involvement was required (prescription, authorization)
Diagnosed which areas of the system needed updating, and which did not
Only after this mapping was reviewed and confirmed did the design work begin.
This process surfaced a principle that stayed with me: when told "there isn't much design work here," that is often the signal to slow down, not speed up. After short diagnosis, what looks like one small screen often contains multiple scenario paths that, if not mapped early, create rework cycles later.
The Solution
Once the mapping was approved:
Dashboard alerts were added to surface the cassette replacement need to medical staff before it became critical
A call-to-action was added on the relevant page within Therapy Manager
Corresponding screens were added to the ReX device itself to complete the cross-system flow
The solution was small in scope and precise in placement, because the diagnosis was done properly first.
The result
A clean, minimal feature addition that fit the existing system without disrupting it. The before/during/after framework used to map this feature became a reusable method for all subsequent feature work: always understand what existed, what changes, and what the downstream state looks like. No rework. No missed scenario paths.
Cross-system UX: Therapy Manager and the ReX device
Therapy Manager didn't operate in isolation. It was paired with the ReX device, a physical medical device used directly by patients. Several features required UX decisions that touched both systems at once.
Designing across both meant holding two user contexts simultaneously: the medical staff using the SaaS portal, and the patient using the physical device. Decisions made on one side had consequences on the other. This cross-system mapping became a recurring discipline throughout the second year, always done in close collaboration with Marcel Caspi (PRODUX).
Final Takeaways
The shift from the first year to the second wasn't just a scope change. It was a thinking change.
Year 1 was about building correctly.
Year 2 was about diagnosing correctly before building.
The principle that came out of this work: never rush to design, even when someone tells you there isn't much design work.
The most expensive UX mistakes aren't in the execution.
They're in the assumptions that were never surfaced before execution started.
Defining what was before, what changes, and what happens after is not a process formality. It's what makes the difference between a feature that fits and one that creates a loop of corrections.
Credits: Marcel Caspi, VP / Senior Product Designer and UX Strategist, PRODUX. This engagement would not have existed without the collaboration we built on the Gyn Tools project that preceded it.