Product Manager exploring how generative AI changes supply chain systems and the PM role itself.

WM2 — An AI-powered warehouse classification system. My learning lab for agentic patterns, semantic search, evaluation frameworks, and AWS deployment.

Burn Rate — A personal finance awareness app that tracks spending trends instead of enforcing budgets. Uses sentiment feedback to learn comfortable spending targets, with a resolution curve showing how burn rate converges over time.

Delivery Radius Calculator — Calculates which zip codes fall within a deliverable radius, reducing a 2-hour manual process to under 5 minutes. React, TypeScript, Python on AWS Lambda.

Flying Stick — A project template for spec-driven development with GitHub Actions CI/CD. The scaffolding I use for cloud-first development with Claude Code — specs in, working software out.

DWCOA Financials — A financial dashboard for a 9-unit condo association. Uses Claude AI to auto-categorize bank transactions. A small project that solves a real problem.

🔧 Agentic systems for supply chain — How to envision, resource, and build adaptive, agentic supply chain systems.

📊 Evaluation frameworks — Writing evaluators into code from the start, not bolting them on later. WM2 is my learning lab for this.

⚙️ Spec-driven development — Specs → plans → tasks → code, managed with GitHub Spec-Kit and Claude Code. I'm interested in how PMs participate as authors, not just reviewers.

🧠 Keeping up — MCP, A2A, front-end design specs, fine-tuning — the landscape shifts fast and I'm trying to stay current.

At Amazon, I lead product for Community Delivery, a program that provides free home delivery from food banks to families in need — 15M+ meals per year across 40+ partnerships.

At Nordstrom, I built an automated inventory repositioning system ($22M pilot EBIT) and a time-series forecasting tool used across Merchandising, Supply Chain, and Store Operations.

💼 LinkedIn | 📧 eve@evehwang.com | 🌐 www.evehwang.com