Algorithms, Explained Like You're Five: Binary Search
"I'm thinking of a number between 1 and 100." You guess 50. "Higher." You guess 75. "Lower." In about seven guesses you've nailed it, because every guess throws away half the numbers left. That's binary search — and it only works because the numbers are in order.
The whole idea in one sentence
On a sorted list, check the middle: if it's too small look only at the right half, if it's too big look only at the left half, and repeat — halving what's left every single time.
Watch it happen
Hit Play. The lit window is what's still in play; each middle-check throws away half of it. Notice how few checks it takes to corner the answer.
Why interviewers love it
It turns a slow "check everything" (O(n)) into a lightning "halve it each step" (O(log n)) — a million items in about twenty checks. The catch they're testing: it only works on sorted data, and the off-by-one bugs (do I use mid+1 or mid? < or <=?) are exactly where people slip. The pattern also hides inside countless "search for the smallest value that works" problems.
Sorted is the price of admission. Pay it once, and every lookup becomes O(log n) — the same halving trick that powers binary search trees.