Inspiration
During our own travel adventures, we frequently encountered the frustration of generic recommendations and the difficulty of meeting like-minded tourists. Existing platforms lacked personalized planning, meaningful connections based on traveler interests, and innovative solutions to common travel problems. Motivated to transform how travel planners and communities function, we envisioned a comprehensive, AI-driven platform that deeply understands travelers and enhances their exploration experiences.
What it does
Our AI-assisted travel planner and social network, "Cluster," provides personalized location recommendations, intelligent image-to-location identification, and powerful tools for connecting travelers based on their preferences, interests, and travel dates. Travelers can upload images of unknown places to identify their locations, use smart searches powered by advanced AI models (Anthropic's Claude API), find travel companions easily, coordinate trips, and engage in communities and live global chat rooms that are actively moderated by AI for safety and engagement.
How we built it
We combined Google's Apps Script and Google Sheets to create a lightweight backend functioning as our data storage and server. Our frontend was developed using HTML, CSS, and JavaScript to power dynamic interactions, responsive designs, and intuitive interfaces. We integrated Anthropic's Claude API to enable advanced AI features, including smart destination matching and image-to-location recognition. Throughout development, we conducted iterative improvements based on testing, feedback, and brainstorming, creating a rapidly evolving and user-centered solution.
Challenges we ran into
One significant issue we encountered was effectively parsing and handling data from different sources (JSON, arrays, URLs) within the limitations of Google Sheets and Apps Script. Additionally, integrating external AI APIs (Claude API by Anthropic) meant managing rate-limiting, response errors, and ensuring seamless frontend-backend communication. There were also UI challenges, especially concerning avatar images being incorrectly displayed as raw URLs or broken icons due to implementation mistakes.
Accomplishments that we're proud of
We're especially proud of successfully integrating an advanced AI-driven system, bringing cutting-edge technologies like image recognition and semantic search directly into travelers' hands. Our work with Google Apps Script demonstrates creative use of readily accessible tools to build a scalable and powerful solution. Moreover, we created a highly interactive and user-friendly frontend, resolved tricky frontend-to-backend data representation issues (like the avatar display), and ensured an intuitive user experience across various devices. Successfully integrating social features with complex backend logic feels particularly rewarding.
What we learned
The development process sharpened our skills in full-stack development, especially around integrating AI-powered functionalities into user-facing products. We learned the importance of effective data management and parsing, robust error-handling, and performant scalability within constrained environments such as Google Sheets. Additionally, through multiple issues and iterations, we improved our debugging skills significantly, understanding common data-flow pitfalls and how to avoid them through better code consistency and data validation.
What's next for AI Assisted Travel Planner and Social Network
Going forward, we plan to expand our AI's capability with more contextual awareness about user preferences, leveraging advanced recommendation algorithms. We're also looking at deeper integration of social interactions, real-time notifications, and potentially incorporating generative AI to help users create more detailed itineraries and personalized suggestions. Additionally, we'll explore using richer datasets and API integrations to continually improve accuracy and the depth of location recognition. Finally, we aim to conduct extensive user testing and gather feedback to prioritize features that genuinely add value to travelers worldwide.
Log in or sign up for Devpost to join the conversation.