We were inspired to build this project, Cultura, as an AI-powered cultural discovery assistant that helps users find personalized recommendations for movies, restaurants, music, activities, and more based on their unique tastes. The idea of combining AI with cultural exploration fascinated me, and We wanted to create a tool that could make discovering new experiences easier and more enjoyable.
Throughout the development process, Me and my team learned a great deal about integrating large language models (LLMs) with real-world APIs to extract user intent and preferences effectively. We also gained experience working with Flask to build a robust backend API, handling data persistence with JSON files, and managing cross-origin requests using Flask-CORS. On the frontend, We focused on creating a clean and user-friendly chat interface using HTML, CSS, and JavaScript, ensuring smooth interaction and responsiveness.
Building this project presented several challenges. One of the main challenges was designing the backend to seamlessly communicate with external AI services like OpenRouter and Qloo while handling potential failures gracefully by providing mock recommendations. Another challenge was managing the conversation history and user preferences efficiently to deliver personalized and context-aware responses. Additionally, ensuring the frontend and backend worked harmoniously required careful API design and testing.
Overall, this project was a rewarding learning experience that deepened my understanding of AI integration, web development, and user experience design. It also highlighted the importance of building resilient systems that can handle real-world uncertainties while delivering valuable functionality to users.
