AIAI Product PrototypesHow to review
Prototype portfolio

Turning AI capabilities into reviewable product experiences

Engineering Lead with 15+ years of experience building mobile, web, and platform products. This portfolio explores how AI capabilities can be translated into practical, understandable, and human-reviewable experiences for business and operational workflows.

All prototypes use fictional data and mock AI services so the product workflow can be validated before model integration.

Prototype

AI Claim Review Workspace

Claims assessors need to review AI-generated summaries with evidence before deciding the next step.

Next.jsTypeScriptTailwindAtomic Components
Prototype

AI Knowledge Answer Reviewer

Operations staff need trustworthy, sourced answers to policy questions with a feedback and escalation loop.

Next.jsTypeScriptTailwindMock AI Adapter
Prototype

AI Assisted Customer Journey Builder

Teams need to turn a broad business goal into an editable draft journey for stakeholder review.

Next.jsTypeScriptTailwindEditable Drafts
Prototype

Operations Insight Dashboard

Operations managers need metrics plus AI insight that links back to the underlying data.

Next.jsTypeScriptRechartsEvidence-linked AI

How to review each prototype

Each prototype runs as a separate Next.js application on its own port. The first two are the highest priority because they most directly demonstrate AI product integration, human review, and regulated-industry thinking.

#1

AI Claim Review Workspace

Runs on port //ai-claim-review.codeduku.com

#2

AI Knowledge Answer Reviewer

Runs on port //ai-knowledge-reviewer.codeduku.com

#3

AI Assisted Customer Journey Builder

Runs on port //ai-journey-builder.codeduku.com

#4

Operations Insight Dashboard

Runs on port //ai-ops-dashboard.codeduku.com

To run locally: npm install then npm run dev in each project folder.