Working in parallel with brand guide development using an agile approach, I designed 5 interconnected platforms that together formed a comprehensive healthcare operations ecosystem. Rapid prototypes were built alongside the design system, ensuring consistency across every surface. All documentation was captured in Confluence for organizational alignment.

Five platforms. One consistent, scalable system.

• Acadia — Claims Operations Platform
  A centralized desktop platform for claims fallout queue management within the ENAC ecosystem. Features included real-time claim status tracking, reason code visibility, LPI clock monitoring, workflow assignment and escalation, informational case acknowledgment, and Denali calculation previews. Built on existing PEAK application themes for consistency across the enterprise suite.


• Provider Management
  A comprehensive provider network management interface supporting 122+ organizations. Designed with filterable data tables across My Providers, My Teams, Approvals, and All Providers views — with at-a-glance KPIs showing new providers added, uploads, and total organizations in network.


• Payer Management
  A guided 7-step payer onboarding flow covering Basic Details, Legal Entity, Address, Contact Details, Bank Details, Exclusions, and Review and Submit — with map integration, address validation, and contextual Notes, Attachments, and Edits panels throughout.


• Member Fund Payment & Financial Reconciliation
  A financial operations platform handling refunds and overpayment workflows. Designed with claims search by Claim Number, TIN, or Member ID, real-time amount validation with error states, and a full audit trail of dates processed and paid.


• AI-Powered Prescription Processing
  A technician intake platform processing 247 prescription documents daily with 89% auto-processing rate and 4.2 minute average turnaround. Features included AI document classification at 94% confidence, OCR extraction with Patient, Prescriber, Drug, and Payer match scores, AI-suggested Next Best Actions, relevant SOP references, and quick actions for finding patients, verifying prescribers, and checking payers.

Across all five platforms I built a scalable design system in Figma leveraging common themes from existing PEAK applications — ensuring a consistent, familiar experience for internal teams across the entire ENAC suite.

At the scale of 400+ features across 5 platforms, the most important design decisions weren't about individual screens — they were about the system-level choices that kept everything consistent, usable, and maintainable.

• Building on PEAK application themes instead of designing from scratch
  Rather than introducing a new visual language, I leveraged existing PEAK application themes as the foundation for all five platforms. This dramatically reduced cognitive load for internal users already familiar with the ecosystem and ensured consistency without sacrificing 18 months of design time on brand decisions.


• Visually distinguishing informational vs actionable claims
  One of the most critical UX decisions in the claims queue was creating a clear visual distinction between cases that required action and cases that were informational only. Without this distinction, operations teams would waste significant time investigating cases that only needed acknowledgment. Clear visual markers resolved this immediately.


• AI confidence scores surfaced at every extraction point
  In the prescription intake platform, I made the decision to show AI confidence scores — 94% for document classification, 92% for patient match, 98% for prescriber match — directly alongside each data point. This gave technicians instant clarity on where to trust the AI and where to verify manually, rather than hiding the model's uncertainty.


• Next Best Action as the core AI interaction model
  Rather than making technicians interpret AI output and decide what to do next, the platform surfaces a prioritized list of Next Best Actions directly in the workflow. This reduced decision fatigue at high document volumes and ensured compliance with prior authorization requirements without relying on technician memory.


• Rapid prototyping in parallel with brand guide development
  Given the 18-month constraint and 400+ feature scope, waiting for a finalized brand guide before designing would have been fatal to the timeline. I made the decision to prototype in parallel, then integrate the design language as it was finalized — a calculated risk that paid off and is now documented in Confluence as an organizational standard.

All five platforms were designed, built, and launched within the 18-month timeline. The following outcomes were achieved at launch:

• 400+ features shipped across 5 platforms
  The full suite launched on schedule — an extraordinary scale of delivery for a single lead designer working within an agile team.


• 89% prescription auto-processing rate
  The AI-powered intake platform achieved an 89% auto-processing rate on prescription documents, dramatically reducing manual technician intervention and bringing average turnaround time to 4.2 minutes.


• 122+ provider organizations onboarded
  The Provider Management platform supported onboarding of 122+ organizations into the network with a consistent, guided experience that replaced fragmented manual processes.


• Centralized claims operations replacing manual fallout handling
  Operations teams gained real-time visibility into claim status, LPI clocks, reason codes, and workflow assignment for the first time — eliminating the guesswork that had previously caused delays and compliance risk.


• SOX compliance tracking integrated into the workflow
  Claim failure status tracking at every step was built directly into the Acadia platform, addressing a significant compliance gap that had previously required manual audit processes.

• Scale requires systems thinking above all else
  At 400+ features, no individual screen decision matters as much as the system-level decisions that govern all of them. The most valuable work I did at TCS wasn't designing any single flow — it was establishing the design language, component library, and documentation standards that made consistency possible at that volume.


• Enterprise users deserve the same design care as consumers
  Internal tools often get treated as second-class design problems. At TCS I treated every claims operations screen and every provider management table with the same rigor I would apply to a consumer product — because the people using these tools every day deserve clarity, efficiency, and an experience that respects their time.


• AI is most valuable when it reduces decisions, not just surfaces data
  The prescription intake platform taught me that AI features succeed when they reduce the number of decisions a user has to make — not just when they surface more information. The Next Best Action model was more impactful than any individual confidence score because it converted AI output into a clear, prioritized action.