Multi-App Enhancement & AI Exploration — MedExpert, Performance & Scalability
Scale: Enterprise-global; multiple interconnected applications serving thousands of users across commercial and medical teams.
Product/System impacted: MedExpert (ME), Smart Targeting (ST), Provider Insights (PI), Patient Disease Insights (PDI), Patient Population Journey (PPJ), and Physician Profile (PP) applications.
Challenge
Main problem: Across six interconnected applications, users experienced inconsistent UX patterns, slow page loads, unreliable search and sorting, and limited data depth. MedExpert specifically needed additional clinical and compliance-related content to become a production-grade tool.
Why it existed: The breadth of the Client platform meant that each app evolved semi-independently, creating UX inconsistencies and technical fragmentation. Rapid feature growth in earlier SOWs created a need for platform-level stabilization.
Impact before project: Inconsistent performance and UX across apps reduced user trust; limited data depth in MedExpert constrained its utility for decision-makers and researchers.
Goals / Success Criteria
Top goals:
- Stabilize and improve UX across all six applications (sorting, search, page load feedback)
- Expand MedExpert with Expert Profile fields, clinical charts, and compliance data
- Investigate and prototype AI/LLM integration to enhance Instant Insights
Key metrics: Reduction in UI/UX inconsistencies; page load performance improvement; new field and chart delivery for MedExpert; feasibility assessment for GPT-powered Instant Insights.
Constraints: Must maintain stability of existing apps throughout; AI exploration must be de-risked and assessed before any production implementation.
Our Role
FLS responsibility: Embedded across the Client’s platform teams, contributing cross-application enhancement, backend refactoring, platform stabilization, Apollo component integration, and AI/LLM feasibility investigation.
Scope & timeline: Multi-sprint delivery across six apps; all sprint objectives met on time.
Teams involved: Project Manager, Business Analyst, 2 Frontend Developers, 2 Backend Developers, QA Analyst.
Approach
Key delivery phases: Platform audit & issue mapping → Cross-app UX fixes → MedExpert feature expansion → Backend refactoring → AI/GPT exploration → QA & stabilization.
Risk-reducing decisions: Centralizing UX improvements at the lookup design level — affecting all apps simultaneously — maximized impact per sprint and reduced future drift. AI exploration was isolated as an investigative workstream, decoupled from production delivery.
Quality assurance: Dedicated backend refactoring for performance; Apollo component integration for consistent front-end data management; comprehensive QA across all six apps.
Solution
Working as an integrated part of the Client’s team, we delivered platform-wide stabilization and feature expansion across six applications, with an AI/GPT integration pilot to enhance Instant Insights.
Core components:
- Cross-app UX improvements: table sorting, page load feedback, search functionality
- MedExpert: new Expert Profile fields, clinical and compliance-related charts and metrics
- Apollo component integration for more flexible, powerful data manipulation
- Backend refactoring for improved overall app performance and scalability
- GPT/LLM investigation for AI-powered Instant Insights enhancement
Differentiator: The decision to redesign application lookups at a shared level — rather than app-by-app — made the entire platform more stable and scalable in one coordinated effort. Working as one team with the Client, we uniquely balanced production delivery with AI exploration within the same sprint cycles.
Technology
Stack: Multi-app frontend + backend; Apollo for data layer; GPT/LLM APIs investigated for Instant Insights integration.
Tooling: Backend profiling for refactoring; Apollo component framework; AI/LLM sandbox testing environment.Accelerators: Shared lookup design system applied across all apps; reusable Apollo components.
Results
Measurable improvements:
- All sprint objectives were met on time across six applications
- Platform-wide lookup design improvement increased stability and scalability across all apps
- MedExpert significantly expanded with new clinical and compliance data dimensions
- Backend refactoring improved overall performance and maintainability
Strongest impact statements:
- A single coordinated design intervention — executed jointly with the Client — improved consistency and scalability across the entire platform, not just one app
- AI/GPT feasibility work positioned the Client for a data-driven decision on LLM integration in future phases
Before / After Snapshot
| Metric | Before | After |
| UX consistency | Fragmented across 6 apps | Unified lookup design applied platform-wide |
| MedExpert data depth | Limited fields & charts | New Expert Profile fields + clinical/compliance charts |
| App performance | Backend performance gaps | Refactored backend; Apollo integration |
| AI capability | No AI in platform | GPT integration investigated; feasibility established |
| Scalability | Per-app fragmentation | Shared component architecture reduces future drift |
AI Enablement
AI used: Yes — investigative/prototype phase.
AI components investigated: Large Language Model (GPT) integration for enhancing Instant Insights — evaluating AI-generated narrative summaries and recommendations based on platform data.
Business problem addressed: Users needed faster, more intuitive data interpretation within the apps. AI-powered Instant Insights could reduce the time from data to decision for commercial and medical users.
AI impact: Feasibility established for production-ready GPT integration. This exploratory work became the foundation for AI-powered features in subsequent phases of the the Client engagement.