The Linear Separability Trap

Senior ML Engineer interview at Stripe, and the interviewer asks:

“Your fraud detection model, a simple linear perceptron, is catching isolated anomalies but completely missing coordinated attacks. Feature A looks safe on its own, and Feature B looks safe on its own, but combined, they scream fraud. A junior engineer suggests throwing 10x more training data at the model. How do you mathematically prove they are wasting time, and what fundamental architectural addition is required to fix it?”

Don’t say: “Let’s switch to AdamW, tune the learning rate, or train for more epochs.”