Inspiration: Traditional credit scoring relies on rigid data. We use LLM for fairer evaluation.
What it does: AI credit scoring via /credit-score, /risk-analysis, /health endpoints.
How we built it: Python FastAPI + LLM scoring algorithm + Pydantic models.
Challenges: LLM accuracy, anti-hallucination, data privacy, low latency.
Accomplishments: Working prototype with contextualized financial insights.
What we learned: LLM interpretability is critical in regulated industries.
Built With: Python, FastAPI, LLM, Pydantic, Uvicorn
Log in or sign up for Devpost to join the conversation.