01 · Life sciences
Making the missing work visible
I used a working prototype to reveal a pre-authoring workflow nobody had formally documented, then turned it into a product module.
01 · Context and ambiguity
The useful problem was underneath the brief.
The brief asked us to improve clinical authoring, but each role described a different failure and the official process missed the work that happened before formal authoring began.
02 · Decisions
I made the system explicit before polishing the surface.
- I mapped the current workflow before defining a future state.
- I built a broad speculative prototype so experts could correct something concrete.
- I consolidated eight content areas into four authoring pillars.
- I carried product intent into code through a project-specific agent harness.
03 · Human and AI boundary
Automation got a lane. Judgment kept ownership.
AI scanned coverage, generated useful defaults, and surfaced questions. Clinical experts retained every consequential study decision.
04 · Working prototype
The artifact became the research surface.
The prototype behaved like the intended product, with representative data, linked states, and enough logic for stakeholders to expose missing dependencies.
05 · Outcome
What changed
Stakeholders recognized their informal one-pager in the prototype. That hidden workflow became the product’s newest module.
Engineering estimated the coded handoff avoided six to eight weeks of delivery effort.