AtlasMe

Description

AtlasMe helps users find their next travel destinations by analyzing their prior trip photos and making personalized recommendations.

Inspiration

Travel planning today often feels uninspired and overwhelming. People tend to revisit the same destinations because discovering new, exciting places takes too much effort. The abundance of options online leads to decision fatigue, and most advice from blogs or friends feels generic and impersonal. Existing planning tools are tedious and lack personalization, making the process feel more like work than an adventure.

We wanted to create something that brings back the joy of discovery, a way for travelers to be inspired by their own memories. That’s why we built AtlasMe, an app that analyzes users’ travel photos to understand what kinds of places they love most. By connecting past experiences to future journeys, AtlasMe helps travelers uncover destinations that feel personally meaningful and exciting.

What it does

AltasMe helps users find new travel destinations based on their previous memories shared through photos. The app analyzes these photos to understand what type of places the user enjoys, then it recommends related destinations they might like to visit next.

Key Features:

AI-powered photo analysis for travel recommendations
An Interactive map with visual destination previews
Personalized destination suggestions based on past trips

How we built it

We built AtlasMe's frontend using React with Next.js, while utilizing AWS backend services.

Challenges we ran into

Integrating the connection between every AWS service was a issue we were having, especially with reworking the lambda functions to fit our specific implementation. Another challenge was reformatting the gemini output into a consistent and accurate output to work with our visuals. We had to adjust a lot of prompting and json formatting to make this work out. A third challenge we ran into was integrating the information with the map. We had to adjust a lot of data types to make the coordinates and routing work with the map correctly.

Accomplishments that we're proud of

Setting up all the AWS services correctly to integrate with each other and give us the quality output that we were looking was rewarding. Also, setting up the recommendations correctly with the AI to give accurate next step trips for users was fulfilling.

What we learned

Vibe coding is good but it is redundent in a lot of things and you spend more time actually fixing the code rather than if you did the whole thing without the ai. It is really good for small things to help you get through something you never have worked on before like the AWS components.

What's next for AtlasMe

Looking ahead, we plan to enhance AtlasMe by adding more features and aim to expand our AI integration to make our recommendations more tailored to your experiences.

Future Implementations:

User Profile: Give an option to upload their preferences for better personalization.
Trip Itinerary: Comes up with an exact trip itinerary that includes cost, significant sights, transportation methods, and accommodation details
App Integration: Give users an app option to make it easier to view their trips from anywhere
Security: Create a login page and authentication for better data protection and user-specific profiles.