Inspiration
The idea for Sift came when we struggled to find a café nearby, highlighting how challenging it is for non-Korean speakers to navigate local maps like KakaoMap. We wanted to simplify this process.
What it does
Sift is an app that helps users find places effortlessly by integrating KakaoMap for searches in Korea and Google Maps for searches abroad. It supports multiple languages and uses LLM to understand natural language queries like “quiet date spot nearby.”
How we built it
We combined KakaoMap and SwiftUI, and we incorporated LLM ChatGPT to process any kind of search query, making it user-friendly for both locals and foreigners.
Challenges we ran into
Integrating two different map systems and ensuring smooth functionality across languages were key challenges, especially while fine-tuning the search algorithms.
Accomplishments that we're proud of
We’re proud of making Sift globally accessible, breaking language barriers, and offering seamless searches with true flexible search queries, tailored for foreigners and locals alike.
What we learned
We gained valuable experience using the ChatGPT API and learned how to effectively integrate KakaoMap into our app. Also Sift allowed us to experiment with asynchronous handling of responses and implement various screen animations, which were exciting challenges for us.
What's next for Sift
Next, we plan to enhance search personalization, improve recommendation accuracy, and integrate more map like GoogleMap to cover additional regions worldwide.
Log in or sign up for Devpost to join the conversation.