Driver Coverage - Compute Geographic Coverage of Active Dashers

Question

## Round 1 - Coding

## Problem

Given a list of active dasher locations and a grid of delivery zones, compute the coverage: the fraction of zones that have at least one dasher within a given radius.

```python
def compute_coverage(
    dashers: list[tuple[float, float]],  # (lat, lon) of each active dasher
    zones: list[dict],                   # [{"id": int, "center": (lat, lon), "radius_km": float}]
    coverage_radius_km: float
) -> float:
    # A zone is "covered" if any dasher is within coverage_radius_km of the zone center
    # Returns fraction of zones covered (0.0 to 1.0)
    ...

def haversine(a: tuple, b: tuple) -> float:
    # Returns great-circle distance in km between two (lat, lon) points
    ...
```

```
Example:
dashers = [(37.77, -122.41), (37.80, -122.40)]
zones = [
    {"id": 1, "center": (37.77, -122.41), "radius_km": 1.0},
    {"id": 2, "center": (37.90, -122.50), "radius_km": 1.0},
]
compute_coverage(dashers, zones, coverage_radius_km=2.0)
# Zone 1: dasher 0 is 0 km away -> covered
# Zone 2: nearest dasher is ~17 km away -> not covered
-> 0.5
```

## Follow-ups

1. Your naive approach is O(D*Z). How would you speed this up for 10K dashers and 5K zones?
2. What spatial index (KD-tree, geohash) would you use for real-time dasher tracking?
3. How would you handle dashers moving in real time - what update frequency is practical?
4. How would you define a coverage SLA (e.g. 95% of zones covered at any time)?

Driver Coverage - Compute Geographic Coverage of Active Dashers

Question Details

Full Details

Topics

More from DoorDash