I build production systems end to end.

Ten of them so far. Job marketplaces, multi-sided logistics, commerce, school management, real-time pipelines.

Each one shipped by one person. That means architecture, data model, authentication, payments, role and permission systems, multi-tenant logic, deployment — and every failure mode in between.

At every organisation I have built for, I was the first engineer they had. There was no existing system to extend and no one to hand the hard part to. Most of these run as internal infrastructure a business depends on daily rather than as public products, which is a different discipline: when it breaks, work stops.

Most of what I build serves users across Uganda, Nigeria and South Africa, where the infrastructure assumptions baked into most software simply do not hold. Intermittent connectivity. Low-end Android. Payment rails that behave nothing like Stripe. Those constraints make for better engineering than scale ever does.