Why should you care?

Swiggy delivery zones on PostgresVerse use polygon GiST to find which zone contains a customer lat/long point.

See it live — copy this example

Run in pgAdmin or psql.

CREATE TABLE delivery_zones (
  zone_id serial PRIMARY KEY,
  zone_name text,
  area polygon
);
CREATE INDEX idx_zones_area_gist ON delivery_zones USING GIST (area);

SELECT zone_name FROM delivery_zones
WHERE area @> point(77.5946, 12.9716);

What happened?

• polygon column stores zone boundaries.
• GiST index accelerates @> point containment — which zone covers Bangalore MG Road coordinates.

Practice next

1. CREATE EXTENSION IF NOT EXISTS postgis is NOT required for built-in polygon/point types.
2. Insert a few zones with polygon values.
3. CREATE GiST index and EXPLAIN the point query.
4. Use daterange column with GiST: WHERE validity && daterange('2026-07-01','2026-07-31');
5. Add EXCLUDE USING GIST to prevent overlapping booking slots.

Remember

GiST handles geometry, ranges, and custom types. @> tests containment for polygons and ranges. Also backs EXCLUDE constraints preventing overlaps.

PostgresVerse geo dispatch

Rider assignment query finds zone in one indexed lookup instead of Python loop.

Outcome: ETA API scales when Bangalore adds 200 micro-zones.