What we build, and how it holds together.

Three groups of services, in the order most clients need them, held together by one discipline. Every engagement is custom — what follows describes what we do, not a menu of packages.

Engineering is not a line item here.

Software engineering is the discipline underneath every service on this page: how the work is specified, reviewed, tested and handed over.

It is also why the services can be combined in one project without the seams showing. Most real projects need several of them at once — an application, the database under it, the integrations around it, and the website in front of it.

Core services

Position in the system: Inner orbit

The systems a business runs on.

Custom software development

Software shaped to how a business actually works, rather than a process reshaped to fit a product.

When a process is specific enough that off-the-shelf products need more workarounds than they save, we build the software around the process instead. We start by mapping how the work really happens, then design the data, the rules and the screens to match it.

Typically involves

  • Mapping the process and where it breaks today
  • Data model and business rules in one inspectable place
  • Interfaces built for the people who do the work

Enterprise applications

Internal platforms with real users, real permissions and a real audit trail.

Applications used every day by teams across an organisation, where roles, approvals and traceability matter as much as the screens. Permissions and audit are designed in from the start, not added when someone asks who changed a record.

Typically involves

  • Role-based access, delegation and approvals
  • Audit trails and reporting that can be defended
  • Connection to the systems already in place

Web application development

Applications that hold up under daily use, on the devices and connections people have.

Browser-based applications for staff, partners or customers, built to stay fast and dependable in real conditions — on phones, on slow connections, in Arabic and in English.

Typically involves

  • Responsive, accessible interfaces
  • Secure sign-in and well-defined APIs
  • Performance tested on real devices

SaaS development

Multi-tenant products built with the operational questions answered before launch.

Products that serve many customers from one platform. Tenancy, onboarding, permissions and monitoring are treated as design decisions on the first day, not problems to solve after the first customers arrive.

Typically involves

  • Multi-tenant architecture and data isolation
  • Tenant onboarding and administration
  • Monitoring and the tools operators need

Specialized solutions

Position in the system: Middle orbit

The parts of an estate most teams would rather not touch.

Oracle APEX solutions

Applications built directly on Oracle data, where the data already lives and the rules already are.

Many companies hold their data in Oracle and still cannot put a usable application in front of the people who need it. APEX is often the shortest honest path from that data to a working system — when it is the right tool. We treat it as one capability among several, connected to the database, integration and modernisation work around it.

Typically involves

  • New applications on existing Oracle estates
  • Modernising applications that work but feel dated
  • REST and ORDS integration with surrounding systems

Database solutions

Schema design, performance work, and migrations that keep their promises about data.

The database is where a system keeps its truth, and often where its slowest problems live. We design schemas, diagnose performance problems that turn out to be structural, and plan migrations so the data arrives complete and verifiable.

Typically involves

  • Schema and data modelling
  • Performance diagnosis and tuning
  • Migrations with verification built in

API & system integrations

Contracts between systems, with the failure cases designed rather than discovered.

Connecting an ERP, a payment provider, a government portal or an internal database. We define the contract first, then build the integration with retries, timeouts, logging and a designed path for when the other side is unavailable.

Typically involves

  • REST APIs and explicit service contracts
  • Retry, timeout and recovery behaviour
  • Logging and an operator view of what failed

Business process automation

Manual steps turned into states a system can hold, audit and recover.

Approvals chased by email, spreadsheets passed between teams, steps that depend on someone remembering. We turn them into workflows with clear states, owners and deadlines, and notifications in the channels people actually read.

Typically involves

  • Workflow and approval design
  • Notifications, reminders and escalation
  • A state history that audit can rely on

Digital experiences

Position in the system: Outer orbit

What people see and touch.

Website design & development

Sites that are fast, accessible and bilingual by construction rather than by retrofit.

Company websites and digital platforms, designed and engineered together. Arabic and right-to-left layouts are designed as first-class experiences, not mirrored at the end.

Typically involves

  • Design and front-end engineering in one team
  • Arabic and English, right-to-left included
  • Performance, accessibility and search foundations

Chatbot development & integration

Conversational interfaces wired to real systems, so an answer is a fact and not a guess.

Assistants for customers or staff that are connected to your actual data and processes, so they can look things up, start a request, and hand the conversation to a person when they should.

Typically involves

  • Connection to business systems and data
  • A clear hand-off to a person
  • Conversation design for the channels you use

A business request, as software

This is the shape of almost every internal system we are asked to build. The interesting part is never the happy path — it is what the system does when something goes wrong.

Select a stage to see what it involves.

Stage 01 / 08

Request

Someone starts a process — a purchase, a claim, an onboarding. It is captured once, with the context needed to act on it, instead of being rebuilt later from email.

A generic enterprise workflow, not a client project.

Not sure which of these you need?

Most projects combine several. Describe the problem in your own words and we will map it to the work it actually requires.