TailorVerseAI

TailorVerseAI is an AI-powered operating system for African fashion businesses. It brings customers, orders, production and everyday operations into one place so a tailoring business can run without scattered notebooks and chat threads.

I came in as the Product Manager on the product team. I ran discovery, set the MVP scope, wrote the PRD and user stories, mapped the flows, designed the mobile-first UX/UI, built prototypes and kept the team moving sprint by sprint.

Product Manager and project manager in one seat. On the product side I owned research, scope, priorities and the experience. On the project side I broke the work into sprints, set timelines and milestones, ran standups, tracked progress and cleared blockers.

I worked with the founder, engineers and design. My job was to keep everyone clear on what we were building, why, and what was due next.

Fashion businesses track measurements, orders, fittings, deliveries and payments across notebooks, phone galleries and chat messages. Details get lost, deadlines slip, and owners have no reliable view of what is due next.

The cost is real: rework on wrong measurements, late deliveries, and money that is hard to trace.

Independent tailors and fashion designers, small fashion studios with a handful of staff, and the clients who order from them.

I ran discovery conversations with fashion business owners to see how they capture measurements, quote work, manage production stages and follow up with clients.

The same three things came up every time: too much to remember, the same details typed again and again, and no single place to see the state of the business.

The opportunity was not another design tool — it was operations. If a tailor could capture a customer once and let the product carry that context through orders, production and delivery, the daily admin work would drop sharply.

Start with the operational spine (customers, measurements, orders, production status), then add AI where it removes typing and decision fatigue instead of adding novelty.

I turned that into a PRD, user stories and a prioritized feature list so scope stayed honest and the team could say no to good ideas that came too early.

In: customer profiles with saved measurements, order creation, production stage tracking, and a dashboard of what is due.

Out for now: payments depth, inventory, marketplace features and anything that needed more than one screen to explain. Cutting those kept the first release small enough to actually ship and test.

Two-week sprints with a clear goal each sprint. I wrote the tickets with acceptance criteria, kept a prioritized backlog, ran short standups and reviewed the build against the flows before anything went to users.

Milestones were tied to user value, not features: capture a customer, run an order end to end, then see the business at a glance.

I mapped the core flows end to end: onboarding a business, adding a customer with measurements, creating an order, moving it through production, and closing it out on delivery.

Designed mobile-first, because most owners run the business from a phone. Big touch targets, short forms, and status you can read at a glance in poor light and on slow data.

AI was applied to the repetitive parts of the job: turning loose notes into structured order details, drafting client messages, and summarising what needs attention today.

Scope pressure. There were more good ideas than sprints. I used the PRD and a simple priority rule — does it help a tailor finish an order faster? — to defend the MVP and park the rest in a visible backlog.

Measurement chaos. Every tailor names and takes measurements differently, so one rigid form kept breaking. We moved to saved measurement sets with custom fields, which cut the arguing and the rework.

Trust in AI. Early users did not want AI deciding things for them. We changed it to draft-and-confirm: AI prepares, the owner approves.

Slow phones and slow data. Screens that looked fine on desktop felt heavy on a mid-range phone, so we cut steps, reduced form length and made status readable without loading extra screens.

Habit, not software. Owners already had a notebook system that worked for them. Onboarding was too long, so we shortened first-run to one thing: add your first customer.

Early users were onboarded directly and watched while they used it. Their feedback changed form length, the production-stage model and the order of the dashboard.

Every round of feedback became sprint tickets, not a document nobody reads.

The product has grown to over 1,200 users. It reached its first 300 users in 2 months during the period I worked on it.

Public milestones shared by the company since then: a web beta with early testers in late 2025, a mobile early-access release, an iPhone app listed in the App Store in March 2026, and later posts putting the product past 1,200 users. Those later numbers are the company's own public figures, not my claim.

On the delivery side: a defined MVP shipped, the core flows working end to end, and a team that always knew the current priority.

Operational products win on less typing, not more features.

Shipping a narrow spine early made every later decision easier to judge.

Onboarding users myself produced sharper insight than any survey.

Most blockers were people problems — clarity, priorities and habits — before they were technical ones.