BarakaAI

• 

Inspiration

Modern credit systems rely heavily on interest-based metrics that often exclude ethically motivated users and underserved communities. As someone who values ethical and Sharia-compliant finance, I was inspired to explore whether AI could assess credit risk without relying on interest at all. BarakaAI was born from the idea that financial responsibility can be measured through behavior, stability, and transparency not interest.

What it does

BarakaAI is an AI-powered, interest-free credit risk assessment system. It evaluates applicants using ethical financial indicators such as income stability, employment experience, credit history length, past repayment behavior, and loan purpose.

The system outputs a Barakah Score (0–100) that represents financial trustworthiness, enabling fairer and more inclusive lending decisions while avoiding interest-based features.

How we built it

Cleaned and preprocessed a real-world financial dataset

Explicitly removed interest-related variables to maintain ethical compliance

Built a machine learning pipeline using scikit-learn

Trained and validated a classification model for loan outcomes

Saved the trained model using joblib

Deployed the system as an interactive Streamlit web app for real-time predictions

All tools used are free and open-source.

Challenges we ran into

Handling messy real-world data with mixed types and inconsistent formats

Preventing data leakage while removing interest-based features

Designing a model that remains accurate without traditional interest signals

Translating technical outputs into a simple, interpretable score for users

Accomplishments that we're proud of

Built a fully working AI system, not just a concept

Successfully removed interest dependency while maintaining performance

Created a live, usable web application

Aligned AI development with ethical and sustainable finance principles

What we learned

Ethical constraints can drive better, more creative AI design

Credit risk can be assessed beyond traditional financial assumptions

Simplicity and transparency matter as much as accuracy

Responsible AI is as much about what you exclude as what you include

What's next for BarakaAI

Add explainable AI (XAI) features for decision transparency

Expand datasets to include underserved populations

Adapt the system for microfinance and SME lending

Partner with ethical finance institutions for real-world pilots

Built With

• joblib
• numpy
• pandas
• python
• scikit-learn
• streamlit