Inspiration
As international students, our first days in the U.S. were overwhelming. We got lost on campus constantly, and even after learning our way around, walking home alone at night still felt nerve-racking. But we soon realized it was never just us: new students, late-shift workers, and commuters all walk home in the dark, and when we asked people at our own hackathon whether they’d walk a seven-minute walk alone at night, they said no.
That experience inspired us to create Buddy Walk — a navigation tool that helps people not only figure out where to go, but also feel more informed and supported while getting there.
What it does
Buddy Walk is a safety-focused walking navigation web app that helps users compare routes based on more than just distance.
Users can enter a destination and view multiple possible walking routes. Each route receives a Safety Score from 0–100 based on factors such as:
- Nearby safe/public places
- Streetlight coverage
- Crime or incident risk
- Connectivity
The score is divided into four categories:
- 0–25: High Risk
- 25–50: Caution
- 50–75: Safer
- 75–100: Very Safe
Buddy Walk also includes functions for situations where the user needs help immediately.
It provides quick access to:
- Find Safe Place — finds nearby public locations where the user may feel safer
- Virtual Call — creates a simulated phone call to make the user appear connected to someone
- SOS / Emergency — provides quick access to emergency assistance
- Movement Safety Monitoring — can detect unusual inactivity during a trip and react accordingly
Buddy Walk requires no account or login, keeping the experience fast and privacy-focused. It proves that even for an app meant to keep people safe, useful software doesn’t need your personal data. The people who most need a safer walk home, like new students or someone out late, shouldn’t have to hand over their identity or their movements just to get one. They open the page, and it works.
How we built it
We built Buddy Walk as a browser-based web application so users can access it easily without installing an app.
The project uses:
- HTML for the website structure
- CSS for the user interface and responsive design
- JavaScript for map interaction, route calculations, safety scoring, location features, and emergency tools
- OpenStreetMap for map data
- Mapping and routing APIs to generate walking routes
- Public geographic data for safety-related information such as streetlights and nearby locations
For the Safety Score, we combine several factors instead of depending on a single data source.
For example:
Safe Places reward routes that pass near useful public locations such as:
- Police stations
- Fire stations
- Hospitals
- Universities
- Libraries
- Restaurants
- Stores
Streetlights help estimate how well-lit a route may be at night.
Crime and incident data reduce the Safety Score when a route passes through areas with higher reported risk.
We also created a prototype connectivity layer to demonstrate how internet availability could become another safety factor. A user with stronger connectivity may have better access to calling, navigation, messaging, and emergency resources.
These values are combined into a final score between 0 and 100, allowing users to compare different routes quickly.
Challenges we ran into
One of our biggest challenges was defining what makes a route "safe."
Safety cannot be represented by only one number. A well-lit street may still have a higher crime risk, while another route may have fewer streetlights but more nearby businesses and public places.
We had to create a scoring system that combines multiple factors while still being simple enough for users to understand.
Another challenge was working with geographic data from different sources. Different datasets have different formats, locations, and levels of detail, so combining them with our routes required additional processing.
Since we built the project during a hackathon, we also had to constantly decide which features were essential and which ones could wait.
Accomplishments that we're proud of
We are proud that Buddy Walk became more than a normal navigation prototype.
Instead of simply telling users how to reach a destination, Buddy Walk provides information about what the journey may be like along the way.
We successfully created:
- Multiple walking route options
- A transparent Safety Score
- Streetlight-based route analysis
- Safe-place detection
- Crime-risk consideration
- A connectivity layer
- Live user location
- An “I Feel Unsafe” emergency menu
- A virtual call feature
- Nearby safe-place search
- Emergency access
- A mobile-friendly browser experience
We are especially proud that the project does not require users to create an account. Someone who feels unsafe should be able to open the website and immediately use the tools they need.
What we learned
This project taught us that building technology for safety is not only a programming problem.
We had to think about how people actually behave when they are nervous, lost, or walking alone.
When someone feels unsafe, they may not want to navigate through several menus or read large amounts of information. Features need to be simple, fast, and immediately understandable.
What's next for Buddy Walk
There are many ways we would like to continue improving Buddy Walk. Our long-term goal is for Buddy Walk to become a safety tool that any campus or city can rely on, and that proves useful software doesn’t need personal data. We want to expand beyond Atlanta, replace our demo connectivity layer with verified network coverage data, let people anonymously flag a dark or unsafe spot without creating an account, and make the Fake Call feel even more real with live, responsive voices. Throughout, the rule stays the same: no accounts, no profiles, nothing stored about you.
Log in or sign up for Devpost to join the conversation.