Welcome
I'm a product designer and builder working across product, technology and design practice.
I design complex software, prototype ideas in working code, and help teams raise their design bar.
Welcome to my digital room
Menu Editor
Hotels were locked into one rigid menu, guests got a broken F&B experience. I led end-to-end design of Menu Manager on Android and Web
9s cut from every guest order, from 24s+
5.6′ to ship a full menu
91% adoption in European markets
Desktop app
Setup happens once on desktop, the payoff shows up in every order taken on the Android POS
Open demoOne product, many sellable forms
Managers were duplicating products per variant one dish, six SKUs for size and flavor. I modeled it as one product with sellable forms instead, so setup happens once.
Configure for the floor, not the CMS
Managers can reorder sections themselves instead of shipping one fixed structure, so setup matches however their venue actually runs.
Fast path when drag isn't
Drag-and-drop handles small menus; a searchable dropdown covers dense ones, so reordering doesn't turn into a bottleneck at scale.
Measuring success
Order speed
Reduce time to complete an order in the POS app
Set up time
Ship a full menu in minutes, not an ops project
Adoption
Flexible enough that properties actually switch over
Menus shaped for the floor
Section order and variants land on POS the way staff actually browse so guest orders take fewer taps
Impact
09s
Cut from every guest order, from over 24s
5.6′
To ship a full menu, end to end
91%
Adoption in European markets
Menu Editor
2025
Guest Checkout
Mews needed US-ready tipping. I shipped it as the entry point, learned in production, then extended into guest checkout.
€24.7K tip revenue generated in the first 20 days
Average tip size rose to 4% from 1.5% baseline
€1.2M in tips processed in the first 35 days
US Tipping
We launched US tipping with presets and service-quality anchors, then used live results to shape how tipping works inside guest checkout
Open demoBefore
Staff had to set a tip on a slider, no presets, and percentages or amounts often had to be calculated manually.
Experiments
Presets and service anchors tested against speed and tip rate
After
Presets and service-quality anchors shown together, guests can pick a service level or a percentage, both in one step
We are right if
Signals
Payment time and tip health, whether tip amounts trend up or down over time.
Proof
Increase tip collection and reduce time to pay in POS app
Learnings
Those patterns define tipping inside guest checkout
Why guest checkout
Now
US tipping was the hard requirement
Bet
Guest self-checkout is the fuller product
Path
Launch, learn, then fold tipping into the guest flow
Vision
One checkout model
Set up in Backoffice, POS app as tracking and fallback for staff and guests scan a QR code to handle bill split, payments and tipping on their devices
Web app
Guest facing, split bills, tip and pay without waiting on staff
POS app
Split bill, tip and pay for assisted checkout
Backoffice configuration
Split bill options, tipping and QR code configuration
Split bills
Scan the QR code in the bill, split the bill equally or by items, tip and pay without waiting on staff
Split by item
Assign individual dishes so each guest pays only for what they ordered
Payment tracking
Follow checkout from scan to payment without leaving the table
Staff takeover
Staff can split and tip on POS when a guest can’t finish from the phone
Impact
4%
Increased avg tip percentage of bill from 1.5% to 4%
€1.2M
Tips processed in the first 35 days
€24.7K
Tip revenue in the first 20 days
Guest Checkout
2025
Business Insurance
Revolut’s insurance quote flow was extensive and unclear. I launched Revolut’s business insurance product from ideation to launch.
-14% drop-off through the quote flow
Weekly quotes grew from 150 to 500+
+150 policies sold a week
Request quote without the Drop-off trap
I aligned web and app surfaces in one onboarding path from request to purchase, post purchase, cancelling flow and edge cases.
Plan history
Active and previous plans with state
Coverage
Extensive coverage documentation readily available
Quote request
Users have the same flow in web and in app
Measuring success
Drop-off
Target reduce drop-off by 20% through the quote path.
Quotes
Higher complete quote requests.
Speed
Fewer steps from request to price.
Impact
-14%
Drop-off reduced vs 20% target.
+500
Weekly quotes, from 150
+150
Policies sold a week
Business Insurance
2024
