Stories in Data

If you have spent any time in tech over the last decade, you know DevOps . It revolutionised how we build things, turning chaotic code deployments into a smooth, automated assembly line. For a long time, the golden rule was simple: If the code passes the automated tests, it’s ready for the world. But then, the world changed. We stopped just writing explicit rules for computers. Instead, we started feeding them massive piles of data and asking them to figure out the rules themselves (machine learning). Shortly after, we started plugging into giant, reasoning foundation models that write essays, summarise documents, and code themselves (generative AI). Traditional software engineering foundations quickly began to crack under the weight. This is the story of how DevOps evolved into MLOps and LLMOps —and why the way we handle lifecycles, data, and system health has changed forever. 1. The Lifecycles: Linear vs. Dynamic Loops In traditional DevOps, the software lifecycle behaves like a wel...