//grind75

Two Pointers — Converging

Sorted array, target sum. Put one pointer at each end and walk them toward each other, deciding which to move by comparing the current sum to the target.

When to reach for it: the input is sorted (or you can sort it), and you’re looking for a pair — or a fixed-size combination — that hits, brackets, or counts against a target sum.

The micropattern:

  1. Sort the array (if it isn’t already).
  2. Place left at the start and right at the end.
  3. Look at nums[left] + nums[right]:
    • too small? move left right to grow the sum.
    • too big? move right left to shrink it.
    • just right? record it and step both inward.

For 3-/4-sum variants you fix the outer element(s) with a loop and run the converging scan on the rest. Work through the problems below in order — each one adds a small twist.

Two Sum II — Input Array Is Sorted

The canonical converging scan: one pass, no extra memory.

Medium LC 167

Container With Most Water

Same converging scan, but move the shorter wall to chase a bigger area.

Medium LC 11

Squares of a Sorted Array

The biggest square is always at one end — fill the result back to front.

Easy LC 977

Boats to Save People

Sort, then pair the lightest with the heaviest that still fits.

Medium LC 881

Intersection of Two Arrays

Two pointers across two sorted arrays, advancing the smaller side.

Easy LC 349

3Sum

Fix one number, then run the converging scan on the rest — plus dedup.

Medium LC 15

3Sum Closest

Same scan, but track the sum nearest the target instead of an exact hit.

Medium LC 16

3Sum Smaller

Counting trick: when the sum is small enough, an entire range of pairs counts at once.

Medium LC 259

4Sum

Two nested anchors, then the same converging scan and dedup.

Medium LC 18