App Development · Nepal
iOS and Android built on a shared core — from a team that ships, operates, and maintains its own live app platform.
Arcetis is a Pokhara-based software company that builds iOS and Android apps for Nepal businesses using a shared-core, cross-platform approach. It built and operates RingAttention, a live restaurant platform whose white-label mobile apps, Stripe payments, and iPad POS are all Arcetis-engineered and piloted with real restaurants.
Most businesses that ask for an app want the outcome an app is associated with — being on the customer's phone, being remembered, taking orders directly — and a mobile website delivers that outcome more cheaply in the majority of cases. An app is genuinely the right build when you need something a browser can't do: reliable push notifications, offline use, home-screen habit for repeat purchase, or connection to payment hardware in a physical location. We apply the same Workaround-Cost Test here as we do for custom software — an app has to beat the cheaper option on real requirements, not on how it sounds.
A SaaS platform for restaurant operations — order and alert management across delivery channels, with integrated payment handling. Every layer below was designed and engineered by Arcetis, and the platform has been piloted with real independent restaurants.
A shared application core configured and branded per restaurant, so each one ships a store-branded app to the App Store and Google Play without a separate codebase per client.
Stripe payment processing including Apple Pay and Google Pay, with persistent cart and guest checkout on the ordering storefront.
In-restaurant point of sale with floor-plan management, per-table QR ordering, kitchen ticket printing, and staff and shift roles.
A customer who orders in the app and later scans a table QR code is recognized as the same person for loyalty and segmentation — one data model behind every surface.
Methodology
A decision framework for choosing between an off-the-shelf CRM/ERP platform and a custom build, based on whether the accumulated cost of workarounds exceeds the cost of a custom system.
1
Not the documented version — how leads actually arrive, how deals actually move, where handoffs actually happen today.
2
What a standard platform would need to bend to fit the actual process, and the ongoing cost of each one.
3
A custom system is only justified when accumulated workaround cost is honestly higher.
4
Most businesses need one or two specific gaps filled, not a fully custom system replacing a standard platform outright.
Cross-platform by default, native where a specific requirement earns it. Flutter is in our core stack, and the shared-core approach — one application core configured per brand — is what makes shipping to both App Store and Google Play affordable. A fully native pair of apps roughly doubles the surface to build and maintain, so it needs a reason beyond preference.
For most Nepal businesses, a fast mobile website or a Progressive Web App is the right answer, and we'll tell you that rather than sell an app. An app earns its cost when you need push notifications, offline behaviour, home-screen habit, hardware access, or in-store payment hardware. If the goal is 'be findable and let people buy', a well-built mobile site does that without an install step.
No flat rate is published, because scope drives it almost entirely: how many user roles, whether payments and third-party integrations are involved, and whether it's one app or a configurable platform. What's fixed is the process — a written scope with a real number, agreed after a discovery call and before any work starts.
Shipped, and we operate it. RingAttention is an Arcetis-built restaurant platform whose white-label mobile app layer runs on a shared core, alongside an iPad POS on Stripe Terminal, Stripe payments with Apple Pay and Google Pay, loyalty, and delivery dispatch. It has been piloted with real independent restaurants — a live product we maintain, not a portfolio screenshot.
Yes — store submission, review handling, and the account setup that goes with it are part of delivery. Getting a build approved is a separate skill from writing the app, and it's a poor place for a client to be left on their own after launch.
Apps are the clearest case for reversibility and ongoing support: operating systems update on their own schedule, and a build that shipped fine last year can break on a new OS release. Maintenance is scoped in the same written agreement as the build rather than sold as a surprise afterwards.