38 points by kylesmith 6 months ago flag hide 14 comments
johnwise 6 months ago next
Great post! I've been looking for a trustworthy personal finance assistant that utilizes AI technology. I am planning to build a similar app, good to know I can experiment with Rust for this use case.
codingcow 6 months ago next
@johnwise I'm excited to follow your progress on building this app. Rust has been gaining popularity lately, and it's great to see more projects adopting this efficient and safe language.
distrosnob 6 months ago prev next
Rust isn't the first language that comes to my mind when I think about building a finance assistant. Have you considered using Python with its extensive libraries and tools for data analysis and machine learning? How does Rust compare?
johnwise 6 months ago next
@distrosnob Rust offers the advantages of low-level memory management, improved performance, and responsiveness, which help ensure data integrity and responsiveness of the app. Also, Rust can write both your frontend and backend, promoting consistency and long-term ease of maintenance.
machinelearner 6 months ago prev next
It's very interesting to see how AI and finance come together. Can the AI make predictions on stock market movements and recommend personalized investment opportunities?
algoquant 6 months ago next
@machinelearner In theory, sophisticated models would be able to make predictions on stock market movements, but it is essential to emphasize the risks and uncertainty of stock markets. To build trust with users, I believe the AI should be aimed towards automating personal budgeting, tracking expenditures, and saving money.
webutils 6 months ago prev next
This can be an exciting project. However, consider data privacy and user consent. Some users might be hesitant to share financial data with an AI-powered platform.
securitips 6 months ago next
@webutils Good call! Data security and privacy are paramount when dealing with personal finance data. Start by getting GDPR, CCPA, and similar certifications. It would help if the application was transparent about what information it tracks, stores, and utilizes for AI processing.
vcalgo 6 months ago prev next
It seems like an ideal approach to give users an option to create custom categories and track expenditures. You may also consider using third-party APIs such as Mint or Plaid to access user bank accounts securely.
johnwise 6 months ago next
@vcalgo That is a fantastic idea to enable custom category creation. I like the idea of using third-party APIs to access user bank accounts since it will save a lot of development time. However, my priority is to assure secure access and data handling.
datasci 6 months ago prev next
How do you plan to train the AI model with user data? Will consent be obtained prior to onboarding new users? How often do you retrain and update the data?
johnwise 6 months ago next
@datasci Consent is crucial, and we'll make it a clear point in the onboarding process. We'd start with preliminary data to seed the AI and reinforce user actions with feedback. Retraining would depend on user base size and frequency but could take place weekly or monthly.
lazycoder 6 months ago prev next
I like this idea, but when do you plan to release an MVP? I would be keen to try it out and see how accurate and helpful the recommendations are.
johnwise 6 months ago next
@lazycoder I appreciate the interest! Our MVP will be available in 8 to 12 weeks, offering basic budgeting, informed goal-setting, and savings recommendations based on user incomes and expenses. Stay tuned for updates.