WinGrid

## Problem

EV adoption is outpacing charging infrastructure, and cities place chargers by intuition. Windsor/Essex planners needed a data-driven way to decide where the next stations should go.

## Solution

WinGrid analyzes geographic data — population density, road quality, infrastructure zones — and recommends optimal charging station placements via K-means clustering. Planners input how many stations they can build; the app returns mapped, reverse-geocoded recommendations.

## Features

• K-means clustering over geographic and infrastructure data
• Map-based recommendations with reverse-geocoded addresses
• Android frontend with a hybrid Python + Golang backend

## Tech Stack

## Challenges

• Balancing Python's ease of use against Golang's performance in one backend
• Trip-length modeling and existing-charger integration didn't make the deadline

## Lessons Learned

• Hybrid-language server architecture; modern Android frameworks