Computer Vision Interview Questions, issue 2, Jan 3, 2026
The Redundant Data Trap
Senior Computer Vision Engineer interview at Google, and the interviewer asks:
“We trained a high-capacity ResNet on 500k images, but it’s still overfitting. My Product Manager wants to spend \$20k to label another 500k random images scraped from the same source. Do you approve the budget?”
Don’t say: “Yes! Deep learning models are data-hungry. To fix high variance, we just need to feed the beast more data.”
Why labeling 500k more images from the same distribution won’t fix overfitting—and how active learning actually moves the decision boundary.
The full answer, with the mechanism and the arithmetic, is for paid subscribers on Substack.