Pathly | Devpost

team photo

“When we navigate, we’re usually asked one question: what’s the fastest way? But for students — especially at night — speed isn’t always the best decision.”

Pathly is a context-aware navigation assistant that helps users make better decisions, not just faster ones.

During the day, Pathly helps students quickly discover the best place right now — whether that’s a quiet library, a nearby café, or a gym — by ranking places using ETA, open status, and live crowd signals, instead of showing a cluttered map.

At night, Pathly shifts focus from speed to confidence. Instead of relying on crime data, we evaluate routes using real-world activity signals like nearby open businesses, transit access, and street density — strong proxies for lighting and human presence.

Users can also set simple comfort preferences, like prioritizing well-lit or busier streets, and Pathly adapts recommendations in real time — without tracking identity or sensitive data.

Unlike Google Maps, which optimizes for speed all day long, Pathly is human-centric: it explains why a place or route is recommended and adapts to time, context, and comfort.

Pathly helps students move through their day — and night — with clarity and confidence.

Challenges we ran into

Integrating multiple live data sources (weather, safety, and environmental data) and ensuring real-time updates was technically complex. We also faced challenges with API rate limits, data consistency, and merging information from different providers. Implementing blockchain wallet support and voice assistant features added extra layers of complexity, especially under tight hackathon time constraints.

Accomplishments that we're proud of

We successfully built a full-stack app that combines live safety data, weather, and AI-powered insights to help users navigate safely. We integrated generative AI (Gemini API) for real-time recommendations, added Solana wallet support, and created a seamless, modern user interface. Our caching and data aggregation pipeline ensures users always get up-to-date, relevant information.

What we learned

We learned how to efficiently aggregate and cache live data from multiple APIs, work with generative AI models for contextual insights, and integrate blockchain features into a web app. We also gained experience in rapid prototyping, team collaboration, and balancing ambitious features with practical implementation.

What's next for Pathly

We plan to expand to more cities, add user-generated safety reports, and further personalize AI recommendations. We aim to improve accessibility, add more languages, and explore partnerships with local organizations to enhance safety data coverage. We’re also interested in deeper blockchain integrations and mobile app development.