Implement Masked Cross-Entropy with Label Smoothing
Company: OpenAI
Role: Machine Learning Engineer
Category: Machine Learning
Difficulty: medium
Interview Round: Technical Screen
Overview: Implement numerically stable masked cross-entropy in NumPy with label smoothing and optional temperature scaling. Validate shapes, dtypes, finite logits, masks, and boundary cases while explaining the loss derivation and gradient.
Read the full OpenAI Machine Learning Engineer interview experience this question came from