FOUNDATIONS • 12 / 397
Start before LLMs. Understand where AI, ML and neural networks fit.
Feature
A feature is an input signal or measurable property used by a model.
Think of it like
Think of Feature as one layer in a very large toolbox: you only need to understand what job that layer performs.
Real life
A phone spam filter, recommendation engine, fraud detector or voice assistant may use Feature somewhere inside.
SRE lens
CPU, memory, error rate and request latency can be features for an anomaly model.
Remember thisA feature is an input signal or measurable property used by a model.