Inspiration
As young adults who are starting to become independent with our own money, we find it difficult to manage our expenses and savings. However, manual expense tracking applications that can help us with our problem are a hassle and often lead to discrepancies and missing entries. Therefore, we decided to create HUAT which streamlines this process and also provides extra features that other automated expense tracking applications do not have.
What it does
Prototype version: Accepts user inputs as a conversational sentence and detect intention of whether the transaction is an income or expenditure, then parsing it into DB for display on frontend
How we built it
React- Springboot- MS SQL DB - with ChatGPT API integration
Challenges we ran into
Time constraints for a fullstack project, Version control during group collaboration, Prompt Engineering ChatGPT API
Accomplishments that we're proud of
Full-stack framework completed, with additional backend methods yet to be implemented on frontend, allowing for easy scaling and expansion into upcoming Dev Version of expanding AI implementation
What we learned
Implementation of AI in daily life, pre-empt prompt engineering to increase accuracy of AI parsing
What's next for Huat Ah
Current Dev Prototype: Able to detect user intention through input prompt and link to backend methods to create income or expenditure transactions Upcoming Dev Version: AI budget forecasting based on upcoming events, past 3 months breakdown of overall budget (Savings, Budget, Income, Expenditure), AI Speech to Text integration to remove user need to type prompts for AI calls. Future Dev Version: Linking with daily use-case through bank-provided API calls (OCBC, DBS etc.) for more seamless user experience
Built With
- chatgptapi
- mssql
- react
- springboot
Log in or sign up for Devpost to join the conversation.