Stories — 22 of them, from 3 companies
Stories from the work.
Short and specific: the problem, what I did, and what came of it. Filter by company or by the kind of problem.
22 stories
- S/01Konin Cloud
Arguing for Next.js, in writing
The initial plan for KoninPay was HTML, Sass and Blade, for a product with merchant and store creation, payments and a wallet that had to keep growing.
Result. The comparison supported the team's technology decision. Product pages became shareable and indexable, and seven live storefronts run on the platform today.
ArchitectureCollaboration→ - S/02Konin Cloud
Test the paths that move money first
In a product with payment workflows and a digital wallet, release bugs hide in the flows nobody clicks through before shipping, and a broken checkout costs far more than a broken button.
Result. Releases were checked where a mistake would cost the most.
Quality→ - S/03Konin Cloud
One click, one payment
A fast double click could send the same payment request twice.
Result. There were no duplicate requests in testing.
QualityCollaboration→ - S/04Konin Cloud
The wallet balance that lagged behind
After a payment, the wallet balance on screen could stay old until a refresh.
Result. The balance on screen matched the real one.
QualityUX & product→ - S/05Konin Cloud
The bug that only Safari saw
A layout broke only in Safari, with a release coming.
Result. Users never saw it.
Quality→ - S/06Konin Cloud
Fetch what you render
A product grid that showed a name, a price and a thumbnail was loading full product objects.
Result. Lighter pages and faster scrolling.
PerformanceCollaboration→ - S/07Konin Cloud
One way to load, fail and retry
Every screen handled loading and errors its own way.
Result. Consistent screens, and new screens became faster to build.
ArchitectureUX & product→ - S/08Konin Cloud
Store setup in steps, not one long form
Merchants had to fill one long form to create a store.
Result. Store setup became easier to finish and easier to support.
UX & product→ - S/09Konin Cloud
Wrong data on screen: frontend or API?
Wrong data on screen raises one question first: is it the frontend or the API?
Result. Issues reached the backend team with evidence attached, not "it's broken".
CollaborationQuality→ - S/10The Sailors
Building the World Cup booking inside the platform
The Sailors needed a FIFA World Cup Qatar 2022 ticket-booking experience for high traffic, under a highly constrained timeline.
Result. Less to build and test under a very tight timeline, and the experience supported about $20,000 in revenue per match event.
ArchitectureOwnership→ - S/11The Sailors
When match tickets opened, traffic spiked
Booking traffic spiked every time World Cup match tickets opened.
Result. Booking stayed usable during the peaks.
Performance→ - S/12The Sailors
From about 55% to 95%+ on Lighthouse
An application was stuck at about 55% on Lighthouse performance.
Result. About 55% to 95%+ on performance.
Performance→ - S/13The Sailors
A fast storefront slows down one feature at a time
New features slowly pulled the storefronts' Lighthouse scores down.
Result. The storefronts stayed at 95%+ across performance, accessibility, SEO, best practices and PWA readiness.
PerformanceQuality→ - S/14The Sailors
Picking a specific seat, online
Customers wanted to choose a specific seat in a cafe, a hall or another venue.
Result. Customers could select specific seats within venues.
UX & productCollaboration→ - S/15The Sailors
One seat changed, the whole map re-rendered
Selecting one seat re-rendered the whole seat map, which was slow on low-end phones.
Result. Selection became smooth on phones.
Performance→ - S/16The Sailors
Server data is not UI state
API data copied into the Redux store went stale and caused bugs.
Result. Fewer stale-data bugs.
Architecture→ - S/17Mars Innovation Technology
Owning Beljwar end to end
Beljwar covered bookings for venues, restaurants, rooms and halls, and its frontend needed one owner from implementation to production.
Result. A booking platform delivered to production, and alongside it 10+ booking and business websites.
Ownership→ - S/18Mars Innovation Technology
The biggest UX win wasn't visual
Date, time, space and confirmation were separate screens between finding a place and reserving it.
Result. A more direct path from discovery to reservation.
UX & product→ - S/19Mars Innovation Technology
Designs are the start of the conversation
20+ Figma and Adobe XD screens had to become production UI.
Result. Responsive, production-ready interfaces built from the designs, improved where it mattered.
UX & productCollaboration→ - S/20Mars Innovation Technology
The states the designs forgot
Designs often had no empty, error or long-text states.
Result. Less rework.
CollaborationUX & product→ - S/21Mars Innovation Technology
Dashboards that keep up as data grows
Large dashboard tables loaded everything at once.
Result. The tables stayed usable as the data grew.
PerformanceArchitecture→ - S/22Mars Innovation Technology
One stylesheet, two directions
The platforms needed right-to-left Arabic and left-to-right English.
Result. No separate RTL copy of the UI to maintain.
Architecture→
The full chapters
Where it happened→