ML Coding: MLP/RNN Forward/Backward Propagation

Problem Overview

This is the 75-minute ML coding round, and it comes in several reported forms, all on the same theme: implement or debug small numerical building blocks in NumPy or PyTorch, then reason about them. Seventeen reports. The best-documented form has four graded parts; the others are listed after it. The…

• 4 candidate-reported follow-ups — the exact probes interviewers asked, with the trigger for each
• The rest of the problem statement — full requirements, constraints, and edge cases
• Approach and trade-offs — what passing candidates did, and the mistakes that sink people