Inspiration
The big impact that can cause a good result. Even though today it may seem line a LLM is great at any problem it was it inspiring to see the challenge that needed a mix of already established algorithms with modern LLM layer.
What it does
Our solution, Polar, is a helpful forensic auditor that helps you detect fradulent transactions that usually go unnoticed.
How we built it
We created a web app that connects to a backend motor that orchestrates calls between the fronted, db and AI workflow.
Challenges we ran into
The intrinsic nature of LLMs and random behavior, API quota, understanding finances laws, data processing, data optimizing, resources optimzing, understading algorithms and having a clear idea of a fraude.
Accomplishments that we're proud of
Have a working app.
What we learned
Have a good team structure.
What's next for Polar
No limits.
Built With
- elevenlabs
- googlegemini
- n8n
- nextjs
- postresql
- python
- tigerdata
- vultr
Log in or sign up for Devpost to join the conversation.