A “Machine Learning Engineer” by professional title but in reality, I tend to focus more on the architectural design of machine learning systems. I started my professional work building a simple linear regression-based algorithm to forecast solar energy usage but over time I moved from seeing the “machine learning” as the only part to one of the moving parts which should go through the same cycle of test, deploy and integrate phase, effectively employing the concept of continuous integration and deployment.