Repeat Rider-Driver Pairs

Find rider and driver pairs that completed at least 3 trips together.

Expected output

rider_id, driver_id, trips. Sorted by trips descending, then rider_id, then driver_id.

Row order is checked. Column names are not checked; numbers are compared to 2 decimals.

Dataset: Ride-share marketplace

Trip requests in four US cities from January to March 2026, with drivers, riders, fares and ratings.

drivers

One row per driver. A driver only serves their own city.

riders

One row per rider account.

trips

One row per trip request, whatever its outcome.

trip_id	INTEGER	primary key, increases with requested_at
rider_id	INTEGER
driver_id	INTEGER	NULL when no driver was assigned
city	TEXT
requested_at	TEXT	YYYY-MM-DD HH:MM:SS (UTC)
dropoff_at	TEXT	NULL unless completed
status	TEXT	completed, cancelled_rider, cancelled_driver or no_driver
distance_km	REAL	estimated route distance
fare	REAL	USD, NULL unless completed
rider_rating	INTEGER	1-5 stars given by the rider, NULL if not rated

Engine: SQLite. Dates are text (YYYY-MM-DD); use strftime(), date() and julianday() for date math.