-
-
MapleBond
Inspiration
The inspiration behind MapleBond came from my own experience as a new immigrant to Canada last year. Settling into a new country involves navigating numerous challenges, from learning the language and finding a job, to securing housing, enrolling children in school, and accessing healthcare. The process of integrating into a new community is overwhelming, and I realized how beneficial it would be to have an AI community assistant to help new immigrants like myself. This led me to the idea of creating MapleBond, an AI-powered assistant designed to support Chinese immigrants in North America with practical, conversational assistance.
What it does
MapleBond is an AI assistant tailored specifically for Chinese immigrants or those interested in immigrating to North America. It provides expert advice and information on various topics including immigration procedures, living arrangements, education, job hunting, and healthcare. By leveraging a large language model (LLM) and specialized data, MapleBond offers precise and convenient information to help users settle down more quickly and comfortably in their new environment.
How we built it
Building MapleBond required developing both the front-end and back-end components of the project. The back-end was developed using Python and the Django framework, while the front-end was built with Next.js and styled with Tailwind CSS. The AI functionality is powered by OpenAI's LLM, and data is stored in a vector database for accurate and context-specific responses. Data is sourced from apps and websites frequently used by new immigrants, such as Xiaohongshu and Zhihu, through web scraping. The project is deployed on Vercel, which offers seamless deployment and integration with GitHub, providing an excellent CI/CD pipeline.
Challenges we ran into
One of the biggest challenges was the need to develop a full-stack project. My background is primarily in back-end development, specifically with C/C++, so I had to quickly learn new technologies for both the front-end and back-end development. Additionally, integrating the various components and ensuring they worked together smoothly was a complex task. Managing data scraping and ensuring the AI provided accurate and relevant responses also required significant effort.
Accomplishments that we're proud of
I'm proud of successfully creating a functional prototype of MapleBond that can assist new immigrants with their queries. Deploying the project on Vercel and seeing it come to life has been incredibly rewarding. The integration of the AI model with the custom data to provide precise and helpful information is a significant accomplishment. I'm also proud of my ability to learn and apply new technologies in a short period, which has broadened my skill set significantly.
What we learned
Throughout this project, I learned a great deal about full-stack development, including how to use Django for back-end development and Next.js for front-end development. I also gained experience with deploying applications on Vercel, managing CI/CD pipelines, and integrating AI models with custom data sources. This project has taught me the importance of seamless integration between different components and the value of user-friendly interfaces in enhancing user experience.
What's next for MapleBond
The MapleBond project is just beginning. The next steps involve:
- Improving UI/UX: Enhancing the user interface to make it more intuitive and user-friendly. This includes summarizing key topics of interest for new immigrants and guiding users to ask relevant questions.
- Expanding Data Sources: Collecting more precise and relevant data to improve the accuracy of the AI assistant's responses. This will involve continuously updating the vector database with new information to better serve the needs of our users.
By focusing on these areas, MapleBond aims to become an indispensable tool for Chinese immigrants, helping them integrate into their new communities more easily and effectively.
Log in or sign up for Devpost to join the conversation.