Roam.AI

Starting up with map that displays curated locations for you.
personalized content(Image) with options to like like/dislike posts and like/dislike creators
personalized content(Blog) with options to like like/dislike posts and like/dislike creators
Personalized AI recommendations based on platform interactions
Personalized AI recommendations based on platform interactions
Leaderboard to add gamified experience
About us page
Our mission and goal
mobile version map view
mobile version contents tab
mobile version of leaderboard tab
mobile version of personalized AI itinerary planner
mobile version of personalized AI itinerary planner

Inspiration

Exploring a city is more than just visiting famous landmarks—it’s about truly experiencing a place the way locals do. Whether you’re fascinated by historical facts, love discovering hidden gems, or just want the best shopping spots, we realized that existing platforms don’t offer a personalized way to explore. That’s why we built Roam.AI—your smart roaming companion that tailors recommendations to your preferences.

What it does

Roam.AI personalizes your exploration experience by curating recommendations based on your interests and interactions. • Home Tab: View the carefully curated location around you. • Content Tab: Discover curated content that adapts to your engagement patterns. • Leaderboard: A gamified system that rewards active contributors, encouraging users to share insights. • Itinerary Tab: Get itinerary suggestions tailored to your preferences, based on platform interactions and available time.

Unlike traditional review-based platforms, Roam.AI eliminates the bias of subjective reviews. What one person dislikes, another might love—so we focus on community-driven discovery.

How we built it

• Frontend: Built with React.js
• Backend: Flask handling data processing and ClaudeAI-powered suggestions.
• Database: SQLite to store user preferences and post content, & SQLAlchemy to integrate it with Flask.
• AI Model: Claude AI analyzes user interactions, user inputs and user's live location to send personalized itinerary.
    • Hosted on AWS

Challenges we ran into

• Building the complete AI pipeline to facilitate personalized recommendations.

Accomplishments that we're proud of

• Successfully integrating Claude AI for dynamic and adaptive recommendations.
• Creating a gamified system that incentivizes community engagement.
• Designing an intuitive and visually appealing UI that enhances user experience.

What we learned

• How user engagement patterns influence recommendation accuracy.
• The importance of context-aware AI in travel and exploration apps.
• The value of community-driven discovery over traditional review-based systems.

What's next for Roam.AI

• We are looking to introduce more social features such as live discussions, events, and meetups, helping users connect with each other and share real-time experiences.

Built With

Submitted to

Startup Week Buildathon

Created by

Worked on system architecture, user profiling pipeline & backend and integration for Google Maps.

Rohan Gore
Seeking ‘26 Full-Time Roles | Ex-AI Intern @ MARV Capital | MS CS @ NYU ’26
I worked on the backend of the application. The backend was developed in Flask and used SQLlite as the database.

Sahil Sarnaik
I worked on integrating Anthropic’s Claude AI, developing the backend, and building the ‘Get Recommendations’ frontend.

YashavikaSingh Singh
I worked on the front end code and designed the UI designs along with the wireframes of each feature and implemented that successfully

Raunak Choudhary