EVSmartCharge
Three charger databases. One privacy-first Android app. Your car's dashboard included.
Android · Kotlin · Jetpack Compose · FastAPI
OVERVIEW
EV charging apps typically require transmitting raw GPS coordinates to a server for charger lookups, building a permanent location history on a third-party server. Existing apps were also single-platform, with no car dashboard integration for the Android Auto display.
An Android app backed by a FastAPI Python backend on Railway. GPS is resolved to a postal code on-device; only the postal code is transmitted. The FastAPI backend aggregates charger data concurrently from three independent sources (Open Charge Map, National Charge Point Registry, OpenStreetMap Overpass), scores chargers with a weighted algorithm, and returns an ordered list. Four Android Auto screens provide a glanceable car dashboard interface.
A GDPR-compliant EV charging assistant with zero GPS coordinates stored server-side, real-time vehicle data via Smartcar, Android Auto integration, and an HMRC mileage export for business journeys.
KEY FEATURES
Privacy-First Location Model
GPS is resolved to a UK postal code on-device using the Android Geocoder. Only the postal code is transmitted, raw coordinates are discarded immediately and never logged. GPS is used locally for navigation handoff only.
TECH STACK
Tap to see the reasoning
BUILD TIMELINE
Privacy Architecture & FastAPI Backend
Designed the postal-code-only data model and on-device GPS resolution flow. Built the FastAPI backend on Railway with PostgreSQL + Redis, implementing the three-source charger aggregation and scoring algorithm.
CHALLENGES & SOLUTIONS
LET'S BUILD SOMETHING TOGETHER
I'm always open to discussing new projects, partnerships, or just a good idea.