Inspiration
- I personally have 7 credit cards, choosing the one which gives maximum cash back is challenging.
What it does
- It will recommend you credit card based on the purchase you are going to make, starting with understanding your credit card statements and recommending how much has been lost in cash backs including offers
How we built it
- Llamaindex for multi agent architecture
- Each AI agent is trained on only 1 credit card data.
Challenges we ran into
- Started with Decompute, took 3 hrs for the code to just respond to hi locally.
- Needs a lot of work on code as well as documentation
Choosing the framework
Accomplishments that we're proud of
- Clarity about the framework to use and the benefits of it.
What we learned
What's next for SwipeSmart
- Finishing the project to run locally and targeting B2B enterprises.
Built With
- llamaindex
- python
- react
- typescript
Log in or sign up for Devpost to join the conversation.