E-commerce · Retail & wholesale
AI customer-service apps in production since January 2026, which replaced a paid writing tool and save about $24,000 a year, alongside AI product imagery, a company knowledge workspace and storefront integrations.
The company AI workspace: one menu for chat, the second brain, the four support apps, marketing and the content engine. An existing capture with empty fields; the team member’s name is replaced and no customer records are displayed.
The context
An online flower retailer has very different workflows across customer support, marketing, storefront and product data. The challenge is connecting useful AI to each team’s actual information and tools.
My contribution
Built and evolved the internal AI platform, the support applications and Chrome extension, the marketing image tools, the company knowledge workspace, storefront integrations and the DIY-kit data migration, from early prototypes in 2025 to production tools in 2026.
The product
Four support apps, shared case history and a Chrome sidepanel, in production since January 2026. They replaced the paid AI writing tool the team used before, which saves about $24,000 a year. See the Customer service AI case study.
Team knowledge around a personal memory, with a chat that chooses which knowledge to read and lists its sources. Two teams’ knowledge is in so far. See the Company Second Brain case study.
Image Studio and Preset Creation for consistent product photos. See the Image Studio case study.
Six months of support conversations classified into customer-question topics and coverage gaps, with a review queue of FAQ and blog drafts built in June 2026. Nothing publishes without a person’s approval. See the Content Engine case study.
The AI Floral Assistant, rush-order requests and Share Your Flair contest entries. See the AI Floral Assistant and Storefront requests case studies.
Farm spreadsheets matched to store products, with recipes, purchase links and 2,789 arrangement plans written to a test copy of the store and checked against their sources. See the DIY kit migration case study.
Implemented supplied wireframes for event planning, flower recipes, presentations and shopping lists. Not launched; work paused when the kit migration took priority. See the Proposal Manager case study.
At a glance / simplified product view
Inside FiftyFlowers
DIY kit data migrationDelivered on a test copy of the store
Company Second BrainMVP · live since July 2026, two of seven team brains filled
Content EngineIn the company AI workspace since June 2026 · nothing published through it yet
Image StudioIn production · released August 2026
Proposal ManagerBuilt from the team’s wireframes · not launched
Customer service AIIn production · since January 2026
Storefront requestsRush form live · contest processing run on 2025 entries
AI Floral AssistantIn the store theme · behind an A/B test since January 2026Engineering choices
Support applications, a browser extension, marketing tools and company knowledge grew into one internal platform while keeping the different workflows each team needs.
Customer-service knowledge supports the operational apps; marketing and engineering knowledge are developing parts of a broader company second brain and 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 tools are at different stages: the support apps and the rush-order form are in production, the shopping assistant ships in the store theme behind the company’s A/B test, Share Your Flair processing has handled every 2025 entry, the knowledge workspace is an MVP, and the kit migration is complete on a test copy of the store.