Application Development

Custom web and mobile application development: real authentication, migrations, tests and a deploy you can trust. Built to hold users, data and money.

Book a call ↗

4.9 on Clutch, 6 reviews4 international awards since 2022Building from Budapest since 2021

What we build

Web applications and mobile apps where something is at stake: bookings, payments, scheduling, customer data. Booking platforms, SaaS products, internal tools that replace a spreadsheet three people are afraid to touch. We have built on Next.js, React, Laravel and React Native since 2021, and more than ten MVPs have gone from idea to launched.

The common thread is that these are applications rather than websites. Something is being recorded, calculated or charged, and being wrong has a cost.

The part that decides whether it works

Not the framework. The data model.

Almost every application that becomes painful to change became painful because the shape of its data was decided in the first week by whoever was fastest, and everything since has been built around that shape. We spend real time on it up front, with you, in language you can follow. It is the least visible part of the project and the one you will feel for years.

How we work

A short discovery, with a written output. What the application has to do, who uses it, what it connects to, and what the first usable version is. You get a document with scope, stages and cost. It is fixed-price and it is yours whether or not we build the thing.

Stages that each end in something real. You use it, we adjust, then we continue. Feedback on software you have actually clicked is worth more than feedback on a specification.

Tests where they earn their place. Not a coverage target, which buys nothing. Tests on the paths that cost you money if they break: signing up, paying, and the two or three actions your product exists to do. Plus a pipeline that runs them on every change and refuses to deploy if they fail.

If you already have something half-built

Increasingly people come to us with an application generated by Lovable, v0, Bolt or Cursor that works in a demo and cannot safely take real customers. That is a specific job with a specific shape, and it has its own page: prototype to production.

What we will tell you honestly

Sometimes the answer is not custom software. If an off-the-shelf product covers eighty per cent of what you need, we will say so, even though configuring it is a much smaller job for us than building from scratch. And if the fastest route to something you can build on is rewriting what you have, that goes in the audit with the reasoning and the cost, before you spend anything hardening the wrong foundation.

How it works

Scoped as something you can stop: Applications go wrong by growing. We break the work into stages that each end with something usable, so you can decide at every boundary whether to continue, change direction, or stop. Anything requiring six months of faith before you see it working is a bet, not a project.
Security done at the start: Authorisation checked on the server and not just hidden in the interface. Database rules written and tested rather than left open. Secrets kept out of anything the browser can read. These are cheap while the application is being built and expensive after it has real users, which is the entire argument for doing them now.
Built for the load you will actually have: Queries written to work at ten thousand rows rather than ten, indexes where they are needed, pagination on anything that lists, and slow work moved out of the request. Then measured rather than assumed, so the number we give you comes from a test.
Migrations, backups, and a staging environment: Your schema under version control, backups we have restored from at least once while you watched, and a staging environment separate from production. None of this demonstrates well and all of it is the difference between a product and a prototype.
Handover as a real option: Documented so your own developers can take it over, or run by us monthly. Both are genuine choices and we will tell you which we think fits. What we will not do is leave you dependent on us because nobody wrote anything down.

Where we have done this

What is and is not included

Included

A written discovery with scope, stages and a fixed price
A data model agreed with you before anything is built
Server-side authorisation, tested database rules, secrets out of the client
Version-controlled migrations, backups restored once while you watch
Tests on the paths that cost money if they break, plus a pipeline
Documentation good enough for another team to take over

Not included

Ongoing hosting and third-party service costs
Content, copy and imagery
App store fees and the review process, though we prepare the submission
Feature work discovered after the scope is agreed, which is quoted first

What we build it with

Next.js
React
React Native
Laravel
Node.js
PostgreSQL
TypeScript
Docker

Questions

The things people ask first.

How do you scope an application?

Into stages that each end with something usable, so you can decide at every boundary whether to continue, change direction or stop. Anything that needs six months of faith before you see it working is a bet rather than a project.

What decides whether it works long term?

Not the framework. The data model. Almost every application that becomes painful to change became painful because the shape of its data was set in week one by whoever was fastest. We spend real time on it up front, with you, in language you can follow.

Can our own developers take it over?

Yes, and that is a genuine option rather than a courtesy. It is documented for a team that has not seen it before, and clients have taken builds in house after launch. The alternative is that we run it monthly, and we will tell you which we think fits.

Do you build mobile apps too?

Yes, with React Native where a single codebase serves both platforms sensibly, which is most business applications. Where it does not, we will say so rather than force it.

What about security?

Authorisation checked on the server rather than hidden in the interface, database rules written and tested, secrets kept out of anything the browser can read. These are cheap while an application is being built and expensive once it has real users, which is the entire argument for doing them now.

We already have something half built. Can you finish it?

Often, yes, and if it came out of Lovable, v0, Bolt or Cursor that is a specific job with a specific shape. It has its own page: prototype to production.

Also in Build

Tell us what you are trying to ship.

Start a conversation ↗