WiseWallet

login
Dashboard
upload

Inspiration

Alois: Coming from a poor family I believe it's very important to track all my finances and optimize my spending as much as possible.

Augusto: Going grocery shopping is always very stressful for me because you never know where you can find the best prices and in the modern world that was very unacceptable for me which is why WiseWallet is such a great idea.

What it does

In theory, our idea was to make an all in one website that would allow users to optimize their spending. It would have features such as receipt upload, potential to allow SSO into their bank accounts. This would allow WiseWallet to parse all the data you enter whether you decide to submit receipts, login to your bank for deeper insights, manually enter receipt data if privacy is your top priority. Given this info, we'd provide insights into what the prices in your area to make sure you're getting the most out of your money. We'd also allow for budget management/allocation with visualization that gets as layered as you'd like it to be. With WiseWallet, our goal would be to make it a one stop shop for all your financial needs.

How we built it

Augusto: I built the UI using Sveltekit, Ts, SCSS, and vite.

Alois: I built the db using mySQL and PHP myadmin. I used python and beautiful soup for the webscraping. For the backend, I used Spring, lombok, hibernate and maven.

Seth: I setup the authentication using Okta with the capability of MFA with twilio. I setup the reception of uploaded files using flask for the webserver. I used OCR Spaces receipt scanner for parsing the image into text and the receipt parsing using python.

Challenges we ran into

Alois: I lost 6hrs due to having a part time job. I also had issues with the webscraping library I was using which caused me to try changing the design of the script. I didn't have time to work on the back end functionality for uploading and storing images.

Augusto: My biggest problem was working out the cookies to avoid users needing to login again after refreshing the page.

Seth: Unfortunately, the OCR API we were using would corrupt the file when it received it or the server would time out. I ran into a time issue. I tried setting up our google cloud environment but none of us had ever used it and we didn't know it took up to 8 hours for DNS records to update so we couldn't verify our domain with google to run our website live. Issues with deploying the data visualization occured but this is what it would be like: https://www.amcharts.com/demos/dragging-pie-slices/