Olytech designs, builds and supports the systems businesses rely on — business software and integrations, applications, applied AI, infrastructure, and connected devices. Based in Cairo and available for projects in Egypt and abroad, we stay accountable from the first look at your operation to the system running in service.
Sunday to Thursday, 09:00–18:00
Olytech exists to replace fragmented, manual operations with working systems, and to keep those systems working. Too much operational work still depends on paper, spreadsheets and software that does not talk to itself. We build systems the people doing the work can actually run.
To be the company a business can bring any system problem to — from the device on the factory floor to the software the office runs on — and get one answer, one plan, and one company accountable for it.
Our engineering practice comes out of more than a decade of automotive software work, where requirements are traced, regressions are caught before release, and what ships has to keep working in the field for a decade. Smaller systems get the same treatment.
Replacing a working system is the most expensive way to improve it. We extend the Odoo instances, databases and hardware that are sound, and say so plainly when they are not.
What gets built, what it covers and the delivery stages are written down and agreed before anything is built. Where something is genuinely uncertain, we say so instead of guessing at it.
A system that nobody maintains stops being an asset. Monitoring, maintenance and incident handling can be part of the engagement, documented for the system rather than left to a general promise.
A young company, so every item here is labelled for what it is — a live product, a working demonstration, a built system, or ongoing practice. Nothing on this page is a claim about a client we have not had.
Bill splitting for tables in Egypt, on the App Store and Google Play. Reads Arabic and English receipts, assigns items across the table, and splits VAT and service proportionally.
See OlyShare →Rooms, guests and daily operations running on Odoo, built as a working demonstration and ready to walk through. Custom Odoo modules are built to order on the same basis.
See the system →Curated AI, DevTools and DevOps news delivered twice a day, filtered by keyword and deduplicated. Anyone can start it; subscribers set their own times, follow single topics and add their own feeds.
See the bot →Live local gold, silver and currency prices for the Egyptian market — by karat and by gram, the guinea, the local ounce and three dollar rates. Free to use, no key, and readable straight from a spreadsheet.
See the API →A catalogue with photos and prices, a till the cashier uses, generated barcodes, printable shelf labels and margin on every item. Clothing, toys and a cafe run side by side on one system.
See the system →Backend services behind applications, Linux and cloud server estates, CI pipelines, deployment and monitoring — the part that keeps a system running after the build is finished.
Take one of them, or the combination a real problem turns out to need.
The unglamorous work where the savings actually are.
AI pointed at a defined job, not a demo.
The screen a person actually uses, and what sits behind it.
Someone has to run it on Monday.
Built to the standard that vehicle software is held to.
When a project crosses several of these at once, it stays one company accountable for the whole system rather than three suppliers negotiating an interface.
Operations these systems are built for, and the problems they solve there. Familiarity with the work, not a claim of past engagements.
Software built to automotive discipline — requirements traced, regressions caught before release, and code expected to survive a decade in the field. The same rigour applied to anything safety-adjacent.
We look at the operation before proposing anything — on site where practical, or remotely through live walkthroughs, existing documents, screens and the numbers you already keep. No proposal is written until we understand how the work is actually done.
What gets built, what it covers and the delivery stages are agreed in writing before work starts. Where something is genuinely uncertain, we say so instead of guessing at it.
We extend the systems you already run rather than replacing them, and put the intelligence where the decision actually gets made.
We put it into service, check it behaves in the real operating environment, and walk the people who will use it through the new process.
Where a system needs looking after, monitoring, maintenance, incident handling and how later improvements get scoped are documented for that engagement.
It is one company, and our engineering came up through automotive software, where both were normal. The embedded side is the reason the AI work holds up — we know what a device can actually do, what a network drops, and what fails at 3am. Most projects need one of these, and that is all we take on.
Tell us what is slowing you down and what you want instead. A sentence is enough — we look at the work first, then quote it.