Batching inputs is how you keep a GPU fed, but the source is an iterator you cannot index or len(). The warm-up that screens whether you can write a lazy generator that handles the ragged final batch.
Updated Aug 2026 · Grounded in real Forward Deployed Engineer interview loops and written to a senior-engineer editorial bar.
Batching inputs is how you keep a GPU fed, but the source is an iterator you cannot index or len(). The warm-up that screens whether you can write a lazy generator that handles the ragged final batch.
Two tells separate a clean answer. First, indexing into the source (data[i:i+n]) when the prompt says iterator, which fails the moment the input is a generator or a 100GB file you cannot slice. Second, dropping or duplicating the last partial batch. The senior instinct is islice or a manual accumulator that yields whatever remains when the source is exhausted, and a parameter for whether the caller wants that partial batch at all.
No comments yet — be the first to share your approach.
