Blog

I have been involved in enterprise Kubernetes adoptions for over seven years — first supporting customers at Docker, then designing platforms as a Solutions Architect at Mirantis. In that time, I have watched organizations succeed, fail, and mostly struggle somewhere in the middle.

Container registries seem like the simplest part of a container platform. They store images. You push to them. You pull from them. What could go wrong? When you are dealing with multiple terabytes of container images across hundreds of repositories, spanning multiple years of organic growth, the answer is: a lot.

Every few years, our industry finds a new label for an old set of problems. Platform engineering is the current label. And like most labels, it is being stretched to mean whatever the speaker wants it to mean.

I am not an AI researcher. I do not build models. I do not train LLMs. What I do is build infrastructure platforms — and I have noticed something interesting over the past eighteen months.

I started my career in a windowless server room in Pune, racking servers for CLSA’s trading platforms. The room was loud, cold, and unforgiving. If a server went down during trading hours, people lost money. Not theoretically — measurably.