I Tried to Turn Accident Data Into a Real-Time Safety App
This project started with a simple thought: if a street has been dangerous for years, why should that knowledge stay hidden in static reports? I wanted to turn that data into something that could actually warn people in real time.
I was working with traffic accident data and seeing the same pattern again and again: certain streets kept showing up. The names changed depending on the city, but the idea stayed the same. Some places were simply riskier than others.
That raised a question I could not ignore.
Why should this information only live in charts, maps, and thesis pages?
If we already know where risk keeps repeating, why not warn people before they get there?That idea became SafeWay.
The goal was not to build a huge platform. It was to build something focused and useful. A mobile app that looks at historical accident patterns, checks the user’s live location, and gives a warning when they approach a risky street segment.
The part I cared about most
I did not want the project to feel theoretical.
I wanted it to feel real.
That meant taking raw accident data, cleaning it, converting coordinates, matching streets properly, classifying risk, and then turning all of that into something a mobile app could actually use.
What made the project interesting
The hardest part was not building a screen. It was making messy, real-world data usable enough for a real product idea.
What I learned
Building SafeWay changed how I think about software projects.
A good app is not just interface and code. Sometimes the real challenge is translating information into action.
In this case, that meant turning old accident records into live warnings.
Quote
The project stopped feeling like a thesis the moment it started feeling useful.