AI-Assisted Design Pipeline: Governed design-to-code workflow for enterprise healthcare modernization
UX Research & Design Strategy Lead
Creating a faster path from design intent to implementation.
I built a VP-backed AI-assisted workflow connecting Figma MCP, GitHub Copilot, and Storybook — reducing design-to-handoff time from 12 weeks to 6–8 weeks.
The goal was not to use AI for its own sake. It was to shorten the path from validated design intent to implementation while keeping human judgment in the loop at every stage.
AI accelerates execution. Humans remain responsible for direction, validation, and approval.
From scattered handoff to governed delivery.
Research and design intent was getting diluted between discovery, prototype, engineering, and implementation. The pipeline connected those stages into one continuous workflow with human review at every gate.
Per 12-week design lead cycle. Reduced buffer between research and engineering handoff.
Senior Director and VP allocated dedicated engineering capacity outside regular roadmap commitments.
Adopted into the team's active delivery practice, not a one-time prototype.
Proven in a sandbox pilot before applying to Treatment Readiness and Demand Forecasting.
The opportunity was to connect research, design, and engineering into one continuous workflow.
The previous model required design to run a full release cycle ahead of engineering — a lead buffer of up to 12 weeks. At each stage, intent was reinterpreted: research findings moved into static mockups, mockups moved into handoff notes, and engineers built from those notes without a clear path back to the original decisions.
The gap between what research reveals and what engineering builds was not a people problem. It was a workflow problem. The pipeline was designed to close that gap — not by removing human judgment, but by reducing the number of times intent had to be re-explained.
The before model shows where intent weakened through sequential handoff. The AI-assisted model shows how research synthesis, live prototyping, design-system mapping, Storybook validation, and GitHub-aligned delivery carry decisions forward with less reinterpretation.
Human review stayed in the loop at every stage.
The workflow did not remove design judgment. It created a faster feedback loop while keeping human review, design-system validation, and enterprise constraints in place.
Human-led decisions
Problem definition, direction-setting, concept evaluation, and final approval stayed with the designer. AI accelerated execution at each stage — it did not make calls.
Enterprise constraints maintained
The pipeline used only approved enterprise tools — Figma, GitHub Copilot, Storybook — operating inside McKesson's security and compliance model, not around it.
Design-system validation
Every component passed through Storybook validation before product integration. Design system decisions were enforced at the code level through Code Connect and Copilot refactoring.
Validated before build
Live prototypes were reviewed by stakeholders and Customer Success before engineering commitment — moving feedback earlier in the cycle, not after build began.
What This Changed
Teams had a clearer path from insight to prototype to implementation-ready direction. The pipeline reduced the design lead buffer by 6–8 weeks per cycle — not by cutting corners, but by connecting the stages that were previously disconnected.
This work was built alongside the Treatment Readiness modernization effort. Finding the hidden work behind treatment readiness →