Inspiration
Via was inspired by a real experience our team had just one week before this hackathon. After prom, our group of seven to eight girls was walking through The Wharf in Washington, DC, around 10 PM on our way to get ice cream. The area was crowded and popular, so we felt completely safe. However, we noticed a group of around 10–15 men slowly moving closer to us as we walked. At first, we thought we were being paranoid, but when we changed directions and increased our pace, they continued following us.
Luckily, we found an open restaurant and were able to wait inside until the group left. Nothing ultimately happened, but the experience left us wondering why getting from one place to another safely often depends on luck.
A bridge exists to help people travel safely between two destinations. Yet when navigating a city, most apps only focus on finding the fastest route, not the safest one. We realized there was a missing bridge between the vast amount of public safety data available and the people who could benefit from it.
That realization became the foundation for Via.
What it does
Via is an AI-powered navigation platform that helps users find the safest route between two locations.
Using over 24,000 Washington, DC crime reports, our system analyzes crime severity, location, frequency, and proximity to roads. Each crime is assigned a danger score from 1–10, with more serious crimes such as homicide receiving higher weights than lower-risk incidents.
Via then evaluates possible routes and generates a safety score for each option and provides the user with the route that had the lowest score (safest).
Instead of simply asking, "What's the fastest way there?", Via answers the question: "What's the safest way there?"
How we built it
We began by collecting and processing publicly available Washington, DC crime data.
Next, we created a danger scoring system that assigns numerical weights to different categories of crimes. We then mapped crime locations onto geographic coordinates and developed an algorithm that calculates risk levels for nearby streets.
Using these risk calculations, we built a route optimization engine that evaluates multiple possible paths and recommends the safest available route.
Finally, we created an interactive web interface that allows users to enter a starting point and destination, visualize risk levels, and view safer navigation options in real time.
Challenges we ran into
One of the biggest challenges was determining how to quantify safety.
Not all crimes have the same impact on personal safety, so we had to carefully design a weighting system that reflected the relative severity of different incidents.
Another challenge was balancing safety and practicality. The absolute safest route is not always realistic if it significantly increases travel time. We spent time designing a system that prioritizes safety while still providing efficient routes.
We also faced challenges processing large datasets and connecting crime data with geographic mapping and route generation.
Accomplishments that we're proud of
We are proud that we transformed more than 24,000 crime reports into actionable safety information that users can understand and use immediately.
We're also proud of creating a project with a meaningful real-world application. Via was inspired by a genuine experience and addresses a problem that many people, especially students, women, tourists, and late-night travelers, face every day.
Most importantly, we successfully connected our solution to the Bridge theme in a way that is both creative and impactful.
What we learned
Through this project, we learned how public datasets can be transformed into useful tools that improve everyday life.
We gained experience working with data processing, route optimization, geographic mapping, and AI-assisted decision-making systems.
We also learned that technology can do more than provide convenience; it can help people feel safer and make more informed decisions.
What's next for Via
In the future, we would like to expand Via beyond historical crime reports by incorporating real-time incident data, community safety reports, street lighting information, and predictive machine learning models.
We also hope to expand to additional cities and create a mobile version of the platform so users can access safer routing information wherever they travel.
Our long-term vision is simple: To make safety a standard part of navigation.
Log in or sign up for Devpost to join the conversation.