I work as a software engineer in the manufacturing industry, where I own internal systems end to end — from data modelling and API design to deployment and on-the-floor support.

My focus is backend engineering for operational software: systems where correctness matters because a person on a shop floor is waiting on the screen, and where a bug stops production rather than annoying a user.

What that has taught me:

• Designing for non-technical users under pressure. The best interface for a factory worker is one they cannot get wrong. That constraint has shaped more architecture decisions than any style guide.
• Integrating with the systems that already exist. Real companies run on ERPs, legacy databases and spreadsheets. Shipping means meeting them there, not replacing them.
• Owning the whole path to production. Linux, Nginx, systemd, TLS, migrations, backups — I deploy and support what I build.

The platform I designed and built to run the production floor of a window and door manufacturer. It tracks every unit from raw material to shipping in real time.

Engineering problems solved on this project:

• A dual-scan workflow that enforces process order. Each workstation scans twice — once to start work, once to release the unit downstream — so no station can hold production hostage and no step can be skipped. Start and finish are stored with separate user and timestamp fields, because the operator who starts a job is often not the one who finishes it.

• One label, two scanning technologies. Every printed label carries a QR Code (structured JSON for phone cameras) and a Code 128 barcode (numeric ID for industrial laser scanners), because workstations have different hardware. A single parser handles both, so the registration flow is identical regardless of device.

• Internal HTTPS with a private certificate authority. Browser cameras require a secure context, and an internal network has no public domain to issue a certificate against. I built a company-local CA, signed the server certificate with it, and shipped a per-platform installation flow — unlocking camera scanning without weakening security.

• A deliberate decision to delete a mobile app. I removed a working Flutter application in favour of a single responsive web screen, after measuring the trade-off: 41 MB less to load on the factory network, no more iOS service-worker caching delaying fixes, and one codebase instead of two.

Languages	Backend & Data	Frontend & Mobile
Infrastructure	Also experienced in Delphi (Pascal) · Windows Server · ERP integrations · REST APIs	Currently exploring System design at scale · Kubernetes · Event-driven architecture · AI-assisted tooling

I started in 2022 as an apprentice, moved through an internship, and now work as a full-time engineer — building and maintaining the systems a manufacturing company runs on.

That path gave me something a textbook cannot: I have written Delphi/Pascal and SQL Server for enterprise operations, and I have shipped modern Python/Django services. I understand both sides of a company's reality — the legacy system that cannot stop running, and the new service that has to earn its place beside it.

Alongside work, I am completing a Computer Science degree.