Geohashing — Part 3: System Design & Interview Mastery

maang.io System Design Series

What is this?

You can now encode a coordinate by hand, explain the Z-curve, and fix the boundary problem. This chapter is where that turns into an offer. When an interviewer sketches "design Uber's nearby-drivers" or "design a proximity feed," they're not testing whether you can recite how bits interleave — they're watching whether you reach for a geohash at the right moment, size the cells correctly, and defend it against a Quadtree when they push back. That judgment is the senior signal; the algorithm is table stakes.

We'll do two things: first see where geohashing lands in real designs with a worked mini-architecture, then run a mock interview — for each question, what they're really asking, a model answer, and the senior line that says you've shipped this for real.

The one-line idea: interviewers use geohashing to test whether you can turn a 2-D "who's near me?" question into a 1-D index scan on infrastructure you already run — and then live honestly with the costs: the boundary problem, uniform cells that hotspot in dense areas, and the coarse-then-fine two-stage filter every real answer needs.