Product Design
Brief Overview -
📌 Role & Background
With deep experience improving Supply Chain Ops and R&D experiences for four of the top Forbes 10 firms in the Lifesciences & Healthcare space in different capacities of building PoCs, RFPs, and MVPs, I’ve spent a lot of time speaking to stakeholders, conducting audits and studying real workflows — the bottlenecks, the silos, and the real human struggles behind every so-called “broken process.” This concept piece isn’t tied to any specific client work, but is born from insights I’ve collected while supporting MVPs and PoCs across high-stakes environments, and also from designing RFPs that converted.
One of the recurring themes I’ve encountered? Complaint management in manufacturing environments—particularly how complex, manual, and fragmented the process still is, even today.
🧾 Complaint Intake
A complaint starts when something goes wrong on the floor—say a vial breaks. Every product is tagged with RFIDs, and tracking it means filling a report. Traditionally? That meant paper forms, rework, and a lot of manual effort. Even with automation systems like Veeva supporting the workers, there's a lot more that needs to be done to make the process smoother
So I reimagined the intake experience.
Using GenAI, the system now assists employees in filling out forms by extracting complaint details from calls, emails, or field logs. It makes capturing context easier—automating tedious parts without stripping away the human nuance.
A key improvement? A final review screen presented as a familiar document-like view. It’s designed so users can cross-check, edit, and feel confident before submitting. It bridges automation with user trust—because people don’t just want speed, they want to be sure.
🚦 Triage
Once submitted, complaints enter the triage phase. This is where most investigators struggle—digging through fragmented info, chasing previous records, and manually prioritizing issues.
So I designed a focused dashboard tailored for pending complaints.
It’s clean, actionable, and filters through noise. Investigators can instantly see key KPIs, review open issues, and search case history—no endless scrolls or hidden menus.
The smart layer? AI-powered suggestions.
Each case is paired with predictive insights: likely issue categories, root cause guesses, and confidence scores—with reasoning included. The goal isn't to replace judgment, but to support it. Investigators can adjust any suggestion, see similar historical complaints, and take action quickly and clearly.
🌱 Learnings & Reflections
Their are some of the key learnings working on this space has taught me:
Stronger the automation layer of system building is, the better the AI/ML gets in building trust with the user groups.
People trust systems they can review, not just automate. The review form wasn’t just a feature—it was the bridge between old habits and new tech.
AI is most powerful when it explains itself. Transparency builds user confidence and improves decisions.
Speed matters, but so does clarity. The goal wasn’t to make everything faster—it was to make decisions easier, more informed, and less frustrating.
Ultimately, this system isn’t just about complaint resolution—it’s about giving people tools that work with them, not just for them.
Whether you’re reaching out for work, want to chat about travel or social impact, or feel like dreaming up ways to make good things happen — I’m all ears. Don’t overthink it, I’d love to hear from you.