Unit III · Searching

Sliding Window

· Step 3/7 · Drill

Drill

Step 3 of 7

Slide the window — reuse the sum

A window slides over a sequence so you re-use work instead of recomputing — turning O(n²) brute force into O(n).

The challenge

Given an array and window size k, return the maximum sum of any contiguous window — without recomputing.

Example

max_window_sum([2, 1, 5, 1, 3, 2], k=3) → 9

You will need to know

Arrays are contiguous in memory — indexing is O(1).
A *window* is a contiguous sub-range `[l..r]`. Fixed-size windows have constant width k.
Sum of a slice can be maintained incrementally: when the window shifts right by one, subtract `nums[l]` and add `nums[r+1]`.
Maintaining a running aggregate turns an O(n·k) brute force into O(n).

window

incremental update

two pointers

Next, your AI tutor will guide you through it — Socratic style.