EZroute: A Concept App
AI-Driven Navigation for NYC Street Fairs

Developed as part of a Human-Centered Design and Innovation course, EzRoute is an AI-powered mobile app concept designed to modernize how New Yorkers navigate and discover the city’s 6,000+ annual street fairs. The project began with the observation that information about these events is often inaccessible or fragmented, creating frustration for drivers and missed opportunities for vendors and pedestrians. Through user research and stakeholder interviews, our team identified a need for a system that proactively alerts commuters about disruptions and connects residents to local events. By integrating real-time traffic data, public event listings, and user preferences through a neural network, EzRoute delivers personalized, context-aware notifications, helping drivers avoid delays and giving pedestrians a curated view of nearby fairs, vendors, and community experiences.

Goals

• Use AI to reimagine how people navigate and discover NYC street events, minimizing frustration while maximizing participation.

• Provide drivers with route-aware event alerts before they begin commuting, without requiring them to manually open a map app.

• Offer pedestrians a personalized event discovery experience, surfacing fairs, farmers' markets, and vendors that align with their interests.

• Support local vendors by giving them a platform to promote offerings and attract foot traffic based on user preferences and location.

• Design an intuitive interface that keeps the experience lightweight, location-aware, and user-configurable.

​

Specs

• AI & ML: Trained a neural network to determine when to send alerts based on location, route history, traffic conditions, and event density

• Data Integration: Aggregated public event data, vendor input, and real-time traffic feeds (e.g., Google Maps API, NYC DOT datasets)

• Mobile Prototyping: Built high-fidelity Figma prototypes for driver and pedestrian modes, including notification flows and route displays

• Backend Design: Outlined logic for route monitoring and time-based alerting using saved routes and custom user settings

• UX Strategy: Developed user profiles and mapped ideal experiences across drivers, pedestrians, and vendors

​

Design process

• Conducted stakeholder research to understand driver frustration with sudden road closures and pedestrian demand for street-level culture.

• Sketched user flows and feature sets for two core modes: commuter alerts and event discovery.

• Used Figma to prototype the app’s core tabs: Home (routes and alerts), Explore (upcoming events), and Community (vendor submissions).

• Developed event cards with live metadata: hours, vendor listings, crowd ratings, and location-based relevance.

• Modeled backend workflows for how user preferences and history shape notifications using a simple rating + feedback loop.

​

Challenges

• Designing for two competing audiences (drivers avoiding events vs. pedestrians seeking them) without cluttering the UI.

• Training a neural network to prioritize meaningful disruptions (not every event justifies a notification).

• Creating a system flexible enough to handle user-defined time windows, routes, and event categories while keeping the interface intuitive.

• Sourcing timely and reliable event data, especially for smaller, community-driven street fairs.

• Ensuring that notifications are both contextually relevant and low-friction in terms of user response.

​

Outcomes

• Delivered a polished mobile prototype demonstrating how AI can improve navigation through urban events, not just around them.

• Created a dual-sided platform that supports both efficient mobility for commuters and economic visibility for local vendors.

• Showcased how a purpose-built tool can outperform Waze or Google Maps in event-heavy, dynamic environments like NYC.

• Developed a scalable framework for expansion into other urban contexts, including bike routes, festivals, and parades.

• Selected as a featured concept in a university innovation program focused on smart cities and sustainable urban mobility.

​

Potential next steps

• Integrate live routing APIs to enable real-time, in-app navigation with overlays for street events.

• Deploy an MVP with basic push notifications and test it with NYC-based users across all three audiences: drivers, pedestrians, and vendors.

• Expand ML model to include event quality prediction based on vendor ratings, crowd size, and social media mentions.

• Develop a vendor dashboard with analytics, profile customization, and direct messaging to interested users.

• Explore partnerships with city agencies to incorporate official road closure data and support public outreach efforts.