Feature Engineering

Feature engineering is the craft of transforming raw data into inputs (features) that make a model's job easier. It covers things like turning a birthdate into an age, combining height and weight into BMI, encoding categories as dummy variables, scaling variables to a common range, or taking the logarithm of a skewed quantity.

It often matters more than the choice of algorithm, because a model can only find patterns that its features make visible — give it the right representation and even a simple regression shines. The discipline is to engineer features using only training data and reproducible logic, so you never accidentally leak future or test information into the inputs.