╔══════════════════════════════════════════════════════════════╗
║   AI × Supply Chain × Product                                ║
║   Building intelligent systems for physical operations       ║
╚══════════════════════════════════════════════════════════════╝

Most AI narratives in supply chain stay at the strategy layer.
I'm interested in what happens at 3am when a sorter jams and the WCS has to decide.

I'm a PM with a background in enterprise warehouse automation — WCS, WES, material handling — now focused on where AI agents meet physical operations. I prototype, analyze, and document AI applications in the distribution center environment: routing intelligence, predictive maintenance, anomaly detection, and human-AI handoff in high-stakes warehouse settings.

This GitHub is my proof-of-work layer.

┌─────────────────────────────────────────────────────┐
│  Where I Work                                       │
│                                                     │
│  Enterprise Supply Chain Software                   │
│  └── Warehouse Control Systems (WCS)                │
│       └── Conveyor / Sortation / Induction          │
│  └── Warehouse Execution Systems (WES)              │
│       └── Labor, Task, Wave Management              │
│  └── SCADA / Industrial Controls (Ignition)         │
│                                                     │
│  Where AI Fits                                      │
│  └── Routing optimization & dynamic re-slotting     │
│  └── Predictive maintenance on material handling    │
│  └── Anomaly detection in throughput data           │
│  └── LLM-powered operator interfaces                │
│  └── AI agents for exception handling               │
└─────────────────────────────────────────────────────┘

The dc-ai-use-cases repo is a structured breakdown of where AI/ML can create real value inside a distribution center — organized by:

• Application area (sortation, induction, labor, maintenance)
• Feasibility tier (Ready Now / 1-2 Years / Research Phase)
• Data requirements (what your WCS/WES needs to already be capturing)
• PM considerations (build vs. buy, integration complexity, change management)

Useful for: AI PMs, supply chain product leaders, and integration engineers evaluating AI roadmaps.

• Prototypes: Small working tools built with Python, Claude API, and occasionally Ignition scripting
• Frameworks: PM decision tools — when to build AI features, how to evaluate routing logic, how to spec LLM interfaces for operators
• Research notes: Public analysis of trends at the intersection of AI and material handling
• Writing cross-posts: Companion repos for my LinkedIn Articles and Medium pieces on AI in supply chain

• Agentic WCS: what does a Claude-powered exception handler look like for a sorter jam?
• Benchmark: rule-based routing vs. ML-assisted routing under variable throughput
• Operator UX for LLM interfaces: how do you design prompts for shift leads, not data scientists?
• AI readiness scoring model for distribution centers