Full-Stack Software Engineer
Full stack, full ownership.
I build ERP, SaaS and booking systems end to end from Hurghada, Egypt architecture, code, deployment, and the maintenance that keeps them running afterwards. AI is part of how I work, not a shortcut: I scope the problem, direct the agent, then read the diff and stand behind what ships.
Live platforms with real users and real money moving through them. Each card shows the request path it actually runs on — hover to trace it.
Distribution ERP covering the full daily cycle: warehouse dispatch, sales, returns, stock closing, cash reconciliation. Custody modelled across five stock locations and six visit outcomes, so goods and cash stay traceable from warehouse to treasury.
Travel booking platform and role-based admin portal for a major agency. DTOs on the .NET side mirror Zod schemas on the front end, so bad data never reaches the database.
Digital ticketing for a Red Sea beach. Takes payment online, then dispatches a QR ticket automatically — no counter, no paper, no queue.
Scan-to-order for restaurants, wired to a live kitchen display over WebSockets. Orders land in the kitchen the moment they are placed. Table turnover up 18%.
Booking platform with an admin dashboard that owns everything: site content, live bookings, pricing tiers and passenger manifests.
Appointment SaaS across six screens. The availability API was read-only, so slot computation moved to the client; tokens stay in HttpOnly cookies behind a proxy route, never in client JS.
CRMs, internal tools and admin dashboards across tourism, medical and legal work. Most of it is client-private — ask and I’ll walk you through it.
Grouped by layer rather than by logo wall. Each one is here because it shipped something on this page.
Part of the normal process, not a shortcut. I scope it, direct it, then review the diff.
Server-rendered by default, so the first paint is real content.
Controller-Service boundaries, validated at the edge.
Schemas that hold under real traffic.
Building it is half the job. Keeping it up is the other half.
Years on the wire before the code — I know what breaks.
Picked per problem, not per habit.
I take a business requirement, not a ticket — and report back in plain language.
Most systems fail before a line of code is written — built for a process nobody actually follows. So I start with the workflow, architect around it, and stay on it after launch.
Sit with the actual workflow before drawing a single screen. Most failed systems were built for a process nobody follows.
Pick the boundaries: what runs on the server, what the client owns, where validation lives. Decided once, up front.
Scope it, direct agentic tools like Claude Code through the implementation, then read the diff and test it. AI moves the typing; the architecture and the accountability stay mine.
Stay on it after launch. Infrastructure, uptime, the awkward calls at 11pm. This is why clients stay.
Two years delivering ERP, CRM and SaaS for real businesses, on top of the infrastructure work that taught me what actually breaks in production.
Hurghada, EG
Software Engineer
Hurghada, EG
Full-Stack Developer
Hurghada, EG
IT Support & Infrastructure Specialist
Hurghada, Egypt
Bachelor of Computer Science
Got a system that needs building, or one that needs rescuing? Tell me what the business actually does and I’ll tell you what it should run on.