PROJECT

Real-Time Geospatial Tracking & Throttling for a Travel App

Engineered a high-fidelity live-map experience with WebSockets, smart throttling, and geospatial validation.

Real-Time Geospatial Tracking & Throttling for a Travel App screenshot

Overview

The client’s live-tracking map was either too chatty—hitting geocoding limits—or too sparse—dots lagging minutes behind. We needed sub-second updates without breaking the bank.

Challenges

  • Balancing real-time update frequency against strict API quotas
  • Avoiding overage fees while keeping UX snappy
  • Ensuring address inputs validated before geocoding

Solutions

  • Switched to WebSockets to stream only changed lat/lon pairs—no full refreshes
  • Built an Elasticsearch-backed geospatial validator to pre-check user inputs
  • Implemented token-bucket throttling to respect Mapbox quotas
  • Cached recent geocoding results in Redis, cutting repeat calls by 60%

Screenshots

Final Notes

Tackling real-time constraints taught me the nuance of efficient API usage and stateful streaming.
  • Location accuracy ↑ 99% (user complaints ↓ 80%)
  • API usage ↓ 60%, avoiding overage fees
  • Real-time updates at 10 Hz with imperceptible lag
  • Geocoding errors ↓ 40%

Interested in updates on new npm releases?

Sign up with your email and get fresh updates as soon as they drop.