A decade of watching things break
— then building my own.
I joined IBM in 2014 as a junior engineer and stayed seven years. The work was unglamorous on the surface — keeping production systems alive for large enterprise customers — and it was the best engineering education I could have asked for. Almost the entire time I was working on the Michelin account, building and operating the systems behind one of the world's largest industrial brands. The recurring lesson, across every incident and every postmortem, was the same: complexity is the enemy of reliability. Clever code fails at 3am. Boring code does not.
In 2021 IBM's infrastructure arm spun out as Kyndryl. I went with it, took on senior IC scope around reliability, observability and platform tooling, and kept working on Michelin — same client, different logo on the badge. Eleven years total at enterprise scale by now. Long enough to know what good infrastructure feels like, and to recognize the absence of it instantly.
In 2025 I began building serious products of my own — a portfolio of focused businesses, each running under its own brand. AURI is reservation software for SPA and wellness operators. Dori Studio is a web studio for B2B and premium B2C. Dataglitch is the third venture on the runway. Same engineering discipline behind each one: same architect end-to-end, no handoffs, no platform tax. More projects will follow the same pattern.
What ties everything together is a single posture: focused scope, observable, predictable, owned end-to-end. The variety — enterprise client, SPA reservation software, marketing websites — is what keeps the craft honest. Same hands, same head, different problems.
Complexity is the enemy of reliability — and the absence of it is the only feature that compounds.