Vectors, matrices, and dot products you can read

A vector is just an ordered list of numbers. In ML work, a row of features is often a vector: recency, frequency, spend, similarity, risk, or whatever the model can inspect.

A dot product multiplies matching positions and adds the results. That makes it a readable scoring recipe: each feature contributes according to its weight.

The builder habit is to keep the feature order visible. A dot product only means something when the numbers line up with the weights you think they line up with.