ISNAD builds applications and AI systems for organisations across the GCC — web, desktop and mobile, agentic workflows, and the security engineering underneath. From a single-branch business to a national enterprise.
Every organisation already runs on systems — order books, spreadsheets, WhatsApp threads, paper forms, software bought for someone else's business. ISNAD builds the version that fits your operation instead: applications, AI, and the engineering underneath, designed around how you actually work.
We are deliberately not a single-product company, and we do not specialise by industry. We specialise in building — web, desktop and mobile applications, AI and agentic systems, integrations, and the security work that keeps all of it defensible. A clinic, a logistics fleet, a manufacturer and a government department need different software; they need the same engineering discipline behind it.
Most problems we are handed need more than one of these — an application that needs an AI layer, or an integration that needs securing before it goes near production. We build them together rather than handing you between vendors.
The interesting work sits where the two hands meet. AI is very good at the parts of a job that are repetitive, high-volume, or easy to forget — and poor at judgement, context, and accountability. We build systems that draw that line deliberately, so your team keeps the decisions and hands over the drudgery.
Web, desktop and mobile built as one system rather than three. Windows and macOS, iOS and Android, and the browser — with Arabic and English side by side where you need it.
Assistants that answer from your own data, and agents that work in the background — drafting, checking, flagging, escalating. Current-generation models, picked per task rather than by fashion.
The unglamorous half: getting systems to talk. Migrations off spreadsheets, connections into what you already own, and reporting that reconciles instead of contradicting itself.
Secure development, code review and vulnerability assessment — on what we build, and on what you already run. Part of delivery, not a certificate bought afterwards.
AI layered onto disorganised data produces confident nonsense. So the order is fixed even when the destination is not: get the records structured, then make them intelligent, then let the system act on its own where that is genuinely safe. You decide where to start and where to stop — plenty of organisations get everything they needed from the first stage.
We replace paper forms, scattered spreadsheets, and WhatsApp threads with unified, mobile-first digital workflows. Every record becomes searchable, traceable, and auditable.
Once data is structured, we layer AI assistants, automated reports, smart suggestions, and natural-language interfaces — bilingual Arabic and English. Teams get answers in seconds, reports in minutes.
AI agents that work continuously in the background — generating briefings before you ask, flagging risks before they manifest, forecasting issues weeks ahead. Humans stay in control; AI handles the heavy lifting.
We are not a single-product company, so there is no module list to fit your business into. These are the functions we most often get asked about — the method behind them is the same, and only the logic changes.
Day-to-day execution
Health, safety, environment
QA/QC, ITP, NCR
Cost, schedule, progress
Vendors, POs, materials
Workforce, training, payroll
AP, AR, reconciliation
Correspondence, drawings
Vehicles, plant, deliveries
Pipeline, tenders, CRM
Permits, licenses, registers
If it's a workflow, we transform it
We do not specialise by sector — we specialise in building. What actually changes between a three-person business and a national enterprise is governance, integration and pace, not whether the work is possible.
One location, a small team, and software bought for somebody else's business. Usually the quickest to feel a difference.
Several branches and a head office that needs one version of the truth. Processes that worked at one site and broke at five.
Procurement, audit trails, data residency, and integration with systems you cannot replace. Phased rollout, documented handover.
These are industries whose workflows we have modelled and can build for. They are grouped for browsing, not for gatekeeping — the last group exists because the list will never be complete.
Orders, inventory, staff scheduling, customer loyalty.
POS systems, supply chain, multi-branch dashboards.
Reservations, housekeeping, guest experience, F&B.
Multi-platform order routing, kitchen ops, delivery tracking.
Inventory, sales tracking, supplier management, loyalty.
Stock control, customer profiles, multi-channel sales.
SKU management, supplier integration, branch dashboards.
Appointments, patient records, billing, lab integrations.
Memberships, class bookings, trainer schedules, payments.
Bookings, customer history, staff commissions, inventory.
Admissions, attendance, fees, parent portals, grading.
Course management, certifications, learner progress, billing.
Production tracking, quality control, maintenance, output.
Vehicle tracking, dispatch, maintenance, fuel logs.
Tenants, contracts, payments, maintenance requests.
Networks, meters, maintenance rounds, outage and consumption reporting.
Plant records, permits to work, shift handover, inspection and asset registers.
Site diaries, HSE observations, progress claims, subcontractor and document control.
Software goes wrong in predictable ways: the scope was never pinned down, the build disappeared for six months, or it shipped and nobody was taught to run it. Each phase below closes one of those gaps.
We watch the work as it is actually done, not as the process document describes it. What comes out is a written scope with the constraints and the things we agreed not to build.
Data model, screens, integrations and permissions, agreed before code. You see the shape of the thing early enough that changing your mind is still cheap.
Delivered in working increments you can open and use, not a status report. Security review runs alongside the build rather than after it.
Training, documentation, source code and credentials — yours. Support continues if you want it, and the system keeps running if you don't.
A first project with a firm you have not worked with before is a bet, and we would rather you placed a small one. Every model below can be entered at pilot size and grown, or kept exactly as it is.
Defined scope. Clear deliverable. Single outcome.
You bring a defined need with a clear endpoint. We scope it together, agree on deliverables and timeline, execute, and hand over. Once complete, the engagement closes — no recurring commitment unless you choose to extend.
Continuous access to expertise.
You retain ISNAD for a regular monthly engagement. We become an extension of your team — available for advice, problem-solving, and execution as needs arise. Your priorities shape what we focus on each month. Predictable access, flexible application.
We build. We run. We hand over.
We design and build the solution. We operate it through the initial period — refining workflows, resolving issues, and proving value in production. Alongside, we train your internal team. When ready, we transfer full ownership. You end up with both a working system and a capable team that owns it.
Long-term partner across multiple engagements.
Beyond a single project, you bring ISNAD into your broader strategy. We work across your portfolio with shared context, continuity, and accountability — scaling our involvement up or down as your pipeline evolves. One trusted partner instead of multiple transactional vendors.
Start small. Prove. Expand at your pace.
We begin with a focused pilot scoped to deliver measurable value on its own. You evaluate the result before any broader commitment. If the pilot succeeds, we expand phase by phase — with go/no-go decisions at every stage. You stay in control of pace, scope, and investment throughout.
You will not be handed between an account manager, a delivery lead and a support desk. Commercially, ISNAD is one accountable point of contact from the first conversation through to handover.
Jamal founded ISNAD and leads its strategy, client engagement and growth, bringing extensive business experience from the UK market into Oman and the wider GCC. Every engagement is scoped, agreed and answered for personally.
Our engineering team works behind the commercial relationship and we do not publish it. What we will put in writing instead is how it works: your code is reviewed before it merges, security assessment runs during the build rather than after, and at handover the source, the credentials and the documentation are transferred to you outright. Nothing we build is designed to make you dependent on us — including on knowing our engineers' names.
ISNAD is being built as a community of contributors rather than a payroll. If you can help us reach clients we cannot reach alone, or build tools we have not built yet, there is a way to work together — and what you get out of it is agreed with you, not assigned to you.
You know an industry, a market or a room we do not. Bring us a real opportunity and take a defined share of what it becomes. Suits people already working in or around a sector who can see where the software should go.
Developers, designers, data and security people. Work on real client systems across web, desktop, mobile and AI — project by project, at whatever depth suits your existing commitments. Show us something you have built.
Legal, finance, translation, training, domain expertise we have never touched. If you can see where you would fit and we cannot, tell us — that conversation costs nothing and has gone somewhere useful before.
How this works. Terms are discussed individually — the arrangement, the commitment and the benefit depend entirely on what you bring and how you want to be involved. Nothing here is a standing offer of employment. Write to us with what you do and what you are looking for.
Apply to Contribute →You do not need a specification, a budget or a decision. A description of the problem is enough to start with — and if we are not the right people for it, we will say so.