Design a location-based radius top-K search
Company: Meta
Role: Software Engineer
Category: System Design
Difficulty: easy
Interview Round: Onsite
Design a location-based search service.
Input:
- `latitude`, `longitude`
- `radius` (meters)
- `K`
Output:
- The top K locations within the radius.
The term “location” is generic and can represent:
- **Static** locations (e.g., businesses/POIs like a Yelp listing)
- **Dynamic** locations (e.g., nearby drivers in a ride-hailing app) that update frequently
Discuss:
- Data model and APIs
- How to index/query efficiently (e.g., geohash, quadtree, or alternatives)
- Trade-offs for static vs dynamic data
- Sharding strategy (including hybrid approaches)
- How you would estimate memory/throughput at a high level
Assume global scale and low-latency queries.
Quick Answer: This question evaluates system-design and backend engineering skills, focusing on geospatial data modeling, indexing and query strategies, sharding and partitioning, handling static versus dynamic location updates, and high-level capacity and latency estimation for throughput-sensitive services.