Companies buy AI.
Teams don't use it.
I close that gap.

Challenge

A multi-location Ben & Jerry's franchise operator (Primo Partners LLC) needed to consolidate inventory, equipment, and infrastructure from four separate retail stores into one centralized operational office in Georgia, as part of a real estate transition and operational downsizing — with commercial kitchen equipment, perishables, and retail fixtures all in play.

Approach

Coordinated two long-haul transport jobs across NC and GA (1,200+ miles round-trip), managing packing, heavy equipment handling, and inventory protection for kiosks, commercial kitchen appliances, signage, and retail fixtures. Ran a 16-ft box truck operation with a crew of 3–7 contractors, working directly with the client and on-site staff to sequence the consolidation safely and on schedule.

Challenge

A family — including materials connected to a family member's published academic and literary work — needed belongings scattered across multiple storage units consolidated into one organized, genuinely accessible system, not just packed into a single space.

Approach

Ran the project across 7 phases spanning residential items, family heirlooms, and archival/literary materials tied to university-affiliated work. Designed a categorization system and floor plan using AI tools (Claude, Gemini, ChatGPT) to organize the collection into 6 distinct, clearly labeled item groups rather than one undifferentiated storage space.

Challenge

180 Milton — a 20-unit luxury student housing property serving the AUC Village community for Morehouse University — needed full move-in-readiness turnover managed under a hard deadline. Twenty units across four wings each carried their own punch list, which expanded mid-project from 5 core items to 15+ categories per unit. Field updates arrived constantly and inconsistently — texted notes, dictated voice memos, photos from a running notes app — sometimes contradicting earlier reports. One unit carried a September 1st student move-in deadline that had to be prioritized above the rest of the property, while the owner needed clear, non-technical status updates without being pulled into job-site noise.

Approach

Applied the Diagnose → Design → Implement → Optimize framework directly to field operations:

• Diagnose: Every raw update — texted punch lists, voice-to-text notes, phone screenshots — was fed into Claude as a single intake point, parsed directly from its native messy format rather than manually re-typed into a spreadsheet.
• Design: Built a standing project data model tracking all 20 units against a growing category list, with priority flags for the hard-deadline unit and access-blocked units tracked separately from routine punch list items.
• Implement: Used Claude to reconcile updates against the standing record in real time — catching regressions and scope changes automatically — generate owner-ready visual status reports, flag discrepancies without silently overwriting prior data, and draft plain-language owner communications. Claude Code and Gemini supported the underlying data pipeline and cross-checking.
• Optimize: Designed the system to absorb new information — corrected quantities, expanded scope, vendor blockers — without losing history, letting completion percentage drop transparently when new scope was discovered rather than hiding it.

Why it matters

Traditional property turnover management for a job this size typically runs on fragmented tracking — a paper list here, a group text there, a spreadsheet that's always a day behind. This engagement delivered the rigor of a large operations team without the headcount, using AI as the actual system of record rather than just a writing tool.