D. MarchamAgricultural Engineering Ltd

Apps for iPhone and Android, and AI that does a real job.

We are a small software company in Reading, England. We design and build native mobile apps, publish them on the App Store and Google Play, and add AI to products and day-to-day work where it clearly saves time.

Email us about a project How a project runs

A plan of four fields seen from above. Each field is one part of our work.
Our work drawn as a field plan. Choose a field to read about it.

What we build

Each field on the plan is one kind of work. Most projects cross two or three of them: an app on both platforms, for example, with an AI feature inside and someone to look after it once it is live.

iPhone and iPad apps

Native apps written in Swift, so they feel right on Apple devices and keep working as iOS changes from year to year.

  • SwiftUI for new work, UIKit where an existing app needs it
  • In-app purchases and subscriptions through StoreKit
  • Widgets, notifications and Sign in with Apple
  • TestFlight builds you can try on your own phone as we go

Android apps

Kotlin apps built with Jetpack Compose and tested on the kind of phones your customers really carry, not only the newest ones.

  • Kotlin and Jetpack Compose
  • Purchases and subscriptions through Google Play Billing
  • Testing on budget handsets as well as flagships
  • Internal and closed testing tracks before a public release

AI in products and operations

We connect language, vision and speech models to your app or to the tools your staff use, and we measure whether they help before anyone depends on them.

  • Assistants that answer from your own documents and quote the source
  • Reading forms, invoices and photos into structured data
  • Speech to text for visit notes and reports
  • Models that run on the phone when data should not leave it

Release and upkeep

Getting an app through review is part of the work. We look after store submissions and keep the app current after launch.

  • App Store and Google Play submissions, and replies to reviewers
  • Updates for each new version of iOS and Android
  • Crash reports and performance monitoring
  • Privacy labels and data safety forms kept accurate

AI, put to work on specific jobs

The AI that pays for itself is usually narrow: one task, done the same way every time, and checked by a person until the results show it can be trusted. These are the kinds of jobs we set up.

  • An engineer dictates a visit report on site, and the app turns it into the form the office already uses.
  • Customer emails are sorted by topic and urgency, each with a suggested reply that a person approves.
  • Staff ask questions in plain English and get answers from your manuals and policies, with the page quoted.
  • Invoices and delivery notes are read and matched to orders without anyone retyping them.
  • Photos taken in the field are checked for the details that matter and flagged when something is missing.

Before it goes live

We start from a set of your real cases and agree in writing what counts as good enough. Personal data goes only to services covered by a data processing agreement, and you can always see why the system suggested something and overrule it.

How a project runs

  1. Survey

    We learn how the work is done today, who will use the app and what success looks like. This usually takes one to two weeks.

  2. Plan

    You receive a written scope, sketches of the main screens, a timetable and a price before any building starts.

  3. Build

    We work in two-week rounds. At the end of each round there is a new build on your phone to try and comment on.

  4. Release

    We prepare the store listings, submit the app and answer the reviewers until it is live for everyone.

  5. Look after

    After launch we fix what users report, update the app for new operating systems and add features as agreed.

Tell us about the job

A few lines are enough to start: what you need, who it is for and when you would like it ready.

Write to support@dmarcham-engineering.com