Bring related tools into one platform.
Evolved support applications, a browser extension, marketing tools and company knowledge into a shared internal platform, while retaining the different workflows each team needs.
E-commerce · Retail & wholesale
AI across the flower business.
AI customer-service apps delivering $24,000 in annual savings, alongside AI image editing and company knowledge.
Internal AI, September 2026. The customer inquiry interface shown with empty fields; no customer records are displayed.
The context
A problem worth solving.
An online flower retailer has very different workflows across customer support, marketing and storefront operations. The challenge is connecting useful AI to each team’s actual information and tools.
My contribution
What I brought to the work.
Built and evolved the internal AI platform, support applications, a Chrome extension, marketing tools and commerce integrations. The work spans full-stack application development, cloud services, product-data workflows and visibility into the company’s AI roadmap.
The product
What it makes possible.
01Customer service+
Four applications for customer questions, reviews, claims and payment disputes, with shared chat, case history and a Chrome extension. The serving layer moved from AWS Lambda to Firebase and Google Cloud. The customer-service apps alone deliver $24,000 in annual savings.
02Company Second Brain+
An umbrella MVP connecting departmental knowledge and tools. Customer Service knowledge, voice and policies support production apps; Marketing demonstrates image-tool guidance and presets; Engineering is a beta used in my own AI-assisted coding.
03Marketing workflows+
Image Studio supports consistent product photos, reusable presets and transparent cutouts. Farm Exports supports catalog-image work. The FAQ & Content Engine helps identify unanswered questions and review content drafts.
04Shopping & operations+
The flower-selection assistant, Share Your Flair wedding submissions and urgent-order requests connect storefront interactions with operational workflows. Earlier implementations and later improvements are distinct parts of the work.
05Proposal Manager+
Implemented supplied wireframes as a tool for event planning, flower recipes, presentations and shopping lists. It was not launched; work paused when DIY Migration took priority.
06DIY Kits Migration+
Tools match farm spreadsheets with store products and transfer kit recipes and purchase links to a test store, with source tracking and checks. Data cleanup and the production handoff remain ongoing.
07Project visibility+
A dedicated FiftyFlowers dashboard in The Record, linked from Internal AI, gives access to the AI roadmap, project milestones, completed work and evidence.
At a glance / simplified product view
- 01Team workflows
- 02Connected AI tools
- 03Human review
Inside the software
See the actual workspace.
Engineering choices
The decisions behind the interface.
Make the rollout boundaries visible.
Customer-service knowledge supports the operational apps. Marketing and engineering knowledge form developing parts of the broader company second brain; they are not presented as equally mature.
The platform combines production workflows with MVP and beta work. The product groups above identify those differences.
Where it stands
The work keeps moving.
Continue the migration cleanup and handoff, and develop the company knowledge workspace beyond its MVP. Individual tools have different levels of maturity; the whole platform is not one finished autonomous system.