Andrew Fahmy · Full-stack engineer · Cairo, Egypt
Your product, shipped end to end.
I design, build, and ship software solo — from database schema to app store — so you get senior-level speed without agency overhead. Currently building Lurnnet, a SaaS platform for educators across Egypt and the GCC.
Fixed priceWeekly demosYou own everything
01Why clients hire me
Most software projects don't fail at the code. They fail in the handoffs — designer to backend, backend to frontend, frontend to DevOps.
I remove the handoffs. One engineer, accountable for the whole system, from the first migration to the app-store release.
- Database schema & migrations
- API design & build
- Web application
- Native mobile app
- CI/CD & infrastructure
- App-store submission
- Monitoring & alerting
- Documentation & handover
02Selected work
A vertical SaaS platform that runs a private teacher's entire practice — classes, assessments, AI-assisted marking, video lessons, and an explainable student-risk engine scoring sixteen live signals.
Architected and built single-handedly: a monorepo serving four user roles across web portals and a native mobile app, with encrypted PII, audit-logged AI, and data residency engineered for Egyptian privacy law ahead of enforcement.
- Backend
- Fastify · PostgreSQL · Redis
- Frontend
- Next.js · React Native (Expo)
- Infra
- AWS Frankfurt · Bunny Stream
- Status
- Pre-launch · founding places open
Tutoring App
A lesson platform for high-school students — quiz-gated content, scheduled exams with instant feedback, and an offline-friendly code-as-balance wallet built around how the teacher actually gets paid. 5,000+ users, 2–3,000 active every month.
02My Joy
A gamified engagement app for a 750+ member community — coins earned through daily check-ins, quizzes, and volunteering, spin-the-wheel rewards, event reservations, and a full admin dashboard. Participation climbed from the first demo.
03EduFlikkr LMS
A full LMS supporting National and IGCSE curricula — AI-driven learning paths, AR-enhanced lessons, and one codebase serving web and mobile. Validated in pilots with educators; its lessons became Lurnnet's foundation.
04EduHire
A recruitment portal connecting schools and teachers — role-based dashboards, one-click applications, advanced filtering, and a clean admin panel. Delivered on schedule; the client praised how intuitive and fast it felt.
03Track record
users on the tutoring app, in production since 2023.
members on My Joy, still maintained and updated since 2024.
commits to Lurnnet, from a team of one.
automated test files guarding Lurnnet.
runbooks written before launch, so operations don't live in one head.
fixes from pre-launch self-audits, each one logged with a date.
Counted, not estimated · Lurnnet figures from its repository, October 2026 · More on Lurnnet →
04About
One engineer, the whole surface.
I'm Andrew — a full-stack engineer in Cairo. Over ten years I've shipped community apps, tutoring platforms, and learning systems used by thousands, rebuilding and refining until the current generation became Lurnnet. That history taught me the thing agencies can't sell you: judgment about what not to build.
When you hire me, you talk to the person writing the code. No account managers, no telephone game between "the team" and the keyboard. Decisions take minutes, not meetings — and nothing gets lost in translation, because there isn't one.
- Based in
- Cairo, Egypt · GMT+2
- Working with
- US · Europe · GCC
- Overlap
- Full day with EU & GCC, 4+ hrs with US East
- Languages
- English · Arabic
- Currently
- Lurnnet — founding places open
05How an engagement works
A written spec, a fixed price
We define the deliverable precisely before any code is written. You get a milestone plan and a fixed price — never an open-ended hourly meter.
Typed end to end, demoed weekly
TypeScript across the stack, tested at the boundaries that matter, with a working demo at every milestone so progress is visible — not promised.
Deployed, documented, handed over
Production deployment, app-store submission where relevant, and documentation your next engineer can actually work from. You own everything.
See it on paper: a scope · a weekly update · a decision record
06How I run production
Checked on every change
Tests, type checks, linting, dependency-vulnerability and secret scanning run automatically on every push, alongside performance and accessibility checks.
Risky features ship switched off
New features go out behind flags and are switched on once they're proven, so a release never has to be all-or-nothing.
Audited, not assumed
Security self-audits before launch with every fix logged, a published vulnerability-disclosure policy, and written incident playbooks, including one for child safeguarding.
Designed in, not bolted on
Personal data is mapped, consent is part of the product rather than a checkbox, and any cross-border data transfer is disclosed, not buried.
Arabic and right-to-left are first-class
Accessibility and right-to-left audits on web and mobile, with every finding tracked to a fix.
Written down, so it doesn't depend on me
Decision records explain why, runbooks explain how, from releases to incident response, so the next engineer can run the system without me in the room.
07What clients say
"Andrew is very proactive — he proposed and built an excellent application to help our youth stay connected with the church and its services. The app is perfect; he delivered everything we wanted, and he keeps following up and updating it as the service needs."
Have a product that needs to ship?
Capacity stays deliberately limited so delivery quality never slips. Q4 2026 has open capacity — let's talk before it's booked.
08What I work with
Product engineering
- —TypeScript, end to end
- —Next.js · React · Tailwind
- —React Native · Expo
- —Zod · TanStack Query
Backend & data
- —Fastify · Node.js
- —PostgreSQL · RLS · pgcrypto
- —Redis caching layers
- —Event-driven scoring engines
Infra & operations
- —AWS · KMS · SES
- —Sentry · PostHog · BetterStack
- —Payment integrations
- —Privacy engineering (PDPL/GDPR)
09What if…
Nothing is stuck in my head
The repository, infrastructure and app-store accounts are in your name from day one, and documentation is written as we build, so another engineer can pick the work up without me.
You keep everything built so far
Payment is split by milestone, and each milestone ends in a working demo. If we stop, you leave with the code, the accounts and the docs. Nothing is held back.
It changes in writing first
New requests are scoped and priced before any code is written. No scope drift without a signed change, and no surprise invoices.
Your next team inherits a system, not a mystery
Typed code, tests at the boundaries that matter, decision records and runbooks, so the engineers you hire next onboard from the docs, not from me.
10Before you ask
Questions, answered.
The things every client asks before the first call — answered here so the call can be about your product instead.
How does pricing work?
Fixed price against a written spec, split across milestones. You see a working demo at each milestone before the next payment is due. No hourly meter, no surprise invoices, no scope drift without a signed change.
How many projects do you take on at once?
A small number, on purpose. Capacity stays capped so every engagement gets senior, hands-on attention from first commit to last — when we talk, you'll know exactly when your build starts.
How do we handle timezones?
I'm in Cairo (GMT+2): a full working-day overlap with Europe and the GCC, and 4+ hours with the US East Coast. The engagement runs async-first — written updates you can read on your schedule, plus a weekly demo call.
What happens after launch?
Every engagement ends with deployment and documentation your next engineer can actually work from. If you'd rather I stay on, we scope a separate support arrangement — but you're never locked in.
What projects fit best?
Full-stack SaaS builds, MVPs with a real deadline, and React Native ports. If you need staff augmentation, I'm not the right fit — and I'll tell you that on the first call, not the last.