CrediCircle

• 

  sample code (back-end)

• 

  sample code (back-end)

Inspiration

In Africa, around 80% of the workforce operates in the informal economy without formal credit histories. Only 23% have access to credit, leaving a $330B financing gap. This gap inspired us to create CrediCircle, an AI-driven platform that empowers underserved communities by providing fair, inclusive, and accessible credit scoring.

What it does

CrediCircle generates AI-powered alternative credit scores using mobile money transactions, call logs, and social trust. It ensures accessibility via USSD, WhatsApp, and mobile app, while integrating blockchain-based trust systems and fraud detection to build secure, transparent, and reliable credit profiles.

How we built it

Backend: Flask + SQLite (scalable to PostgreSQL/MongoDB). AI Models: Python (Scikit-learn/TensorFlow) to analyze alternative data. Blockchain: Solidity prototype for decentralized vouching. Accessibility: USSD via Africa’s Talking, WhatsApp chatbot with Twilio, and React Native app. Fraud Prevention: AI anomaly detection to flag suspicious patterns.

Challenges we ran into

*The lack of sufficient data for developing credit models in informal markets. *Guaranteeing equity in artificial intelligence to prevent bias. *Combining various platforms (USSD, WhatsApp, blockchain). *Creating a system that is both scalable and secure.

Accomplishments that we're proud of

1. Established accessible entry points for both smartphones and feature phones.
2. Engineered a trust-based system that fosters community empowerment.
3. Formulated a solution that aligns with the objectives of financial inclusion and economic growth.

What we learned

1. Strategies for integrating FinTech, artificial intelligence, and blockchain technology to achieve social impact.
2. The significance of creating designs that prioritize user accessibility.
3. The difficulties associated with tackling bias and ensuring fairness in credit scoring systems.
4. Developing modular solutions that are capable of scaling across various African markets.

What's next for CrediCircle

*Incorporate partnerships that provide real-world financial data to enhance and develop AI models. *Enhance fraud detection capabilities through the implementation of sophisticated anomaly detection techniques. *Develop the mobile application into a comprehensive Minimum Viable Product (MVP) for initial testing. *Work in conjunction with microfinance institutions and banks to implement solutions on a large scale. *Expand operations throughout Africa to address the $330 billion credit deficit and foster financial inclusion.

Built With

• africa?s-talking-api
• flask-+-twilio
• mongodb
• native
• node.js-api
• postgresql
• python-(scikit-learn
• react
• solidity
• tensorflow)