The idea behind developing the AI Expense Manager application was based on the common problem of people trying to control their personal expenses. For many individuals, it is quite difficult to monitor their spending, plan their expenditure and even save for their future endeavors. The goal was to design a solution that would not only make this process easier but also smarter. Thus, in our case, the use of AI is to help users make wiser financial decisions based on the analysis of the data provided. From the process of implementing this project, we gained a lot of insight on personal finance management and how AI can be used to solve such issues. We learnt about the user behavior, the aspect of data privacy and the need to develop an interface that would be easy to navigate. Also, we expanded our knowledge of machine learning algorithms, natural language processing, and how those technologies work in practice.
The AI Expense Manager was developed with the help of modern technologies and following the industry standards for software creation. The process included need analysis to determine users’ requirements, analysis of existing tools, and architecture of the application. During design, the emphasis was made on the interfaces and how to make the user experience as easy as possible. Development comprised of code writing, incorporating AI models and establishing databases. Thorough testing was done on the application to make sure that it performs as required and has no defects. Last but not least, we deployed the application, thus opening it to the users.
It was not without its difficulties to build the AI Expense Manager. The challenges that we faced included data privacy, creating an AI that is capable of giving good recommendations, creating an interface that will be adopted by the users, and the ability to integrate with different institutions to ensure that the data is synchronized. For coding, the project utilized Python and JavaScript and the frameworks of Django and React The platforms that were used include AWS and Heroku while the cloud service was AWS S3 and AWS Lambda The database used was PostgreSQL and the financial data integration was from Plaid API For the AI part, OpenAI was used.
Log in or sign up for Devpost to join the conversation.