Travel · United Arab Emirates · 2026
A halal travel platform that books hotels, holidays, cruises and activities
Five booking and discovery journeys in one product, for travellers who need every stay and meal to meet halal standards.
The challenge
Halal travel is normally assembled by hand. A traveller books a hotel on one site, then spends an evening cross-checking whether the kitchen is certified, whether there is a prayer space, and whether the excursions nearby are family-appropriate. Halal Explore set out to collapse that research into a single booking flow, which meant one product had to carry five different journeys at once.
Our approach
We built the front end as a Next.js application against a separate booking API, so the two can be developed and scaled independently and the catalogue can grow without a front-end release. Hotels, holidays, cruises, activities and restaurants each needed their own search, filter and detail patterns, so we designed one shared component vocabulary and specialised it per vertical rather than building five separate interfaces that would drift apart.
What we delivered
- Booking and discovery flows for hotels, holidays, cruises, activities and restaurants
- Next.js front end against a decoupled API, so catalogue changes need no redeploy
- One shared component library specialised per vertical instead of five parallel UIs
- Destination and city landing pages built for organic search
- Editorial section for halal travel guides, tips and stories
- Partner and network integrations with hospitality and travel brands
More work
Another one.
SampleLogistics · Bangladesh · 2025
A Dhaka-based logistics operator
Replacing a spreadsheet dispatch process with a live operations platform
A fleet of 200 vehicles was being dispatched from a shared spreadsheet. We built the platform that replaced it.
- Dispatch time per shipment
- -62%
- Status phone calls
- -80%
- Depots supported
- 4 to 11
SampleRegTech · Sweden · 2025
A Nordic compliance software company
An AI copilot that reads regulation so analysts do not have to
Analysts spent their days searching a 40,000-document regulation corpus. We gave them an assistant that cites its sources.
- Median research time
- 4h to 20m
- Citation accuracy on eval set
- 98.6%
- Cost per answered query
- $0.04
