I pay my own rent for my apartment and there is no option to auto-pay. That got me thinking about this concept in the first place and it developed into API I created.

What it does

Monitor client transactions, and use machine learning to identify patterns that would suggest a possible anomalous situation in which either the client or financial institution should receive a warning.

How I built it

* Use Capital One’s Nessie API to simulate a Bank
* Use Google Cloud for creating the API, with DB
* Use Google Cloud to create a simple website simulating a bank’s customer portal.
* This website will make API calls to both Nessie and The Scrutinizer to identify and display warnings to the customers.
* The Scrutinizer will utilize Google Cloud’s machine learning service.
* Use Postman to manually make API calls.
* Store the project in GitHub.

Challenges I ran into

Time was tight. Didn't implement machine learning, instead used analyze function to detect the patterns. In this case, paying rent.

Accomplishments that I'm proud of

Got further than expected. There is a working demo, website, and was able to successfully use Nessie API and learn a lot from this experience.

What I learned

Learned how to create an and implement API. Learned how to use google cloud.

What's next for The Scrutinizer

Implement machine learning and identify other patterns besides paying rent

Share this project: