August 20, 2026

MobiFin AI

MobiFin homepage with the tagline 'Turn Financial Activity Into Financial Access' and a Get Started as an Agent button.

MobiFin is an AI-powered platform that turns everyday Mobile Money (MoMo) activity into financial intelligence — for the agents who run it, and for the customers it serves. Built by Team MobiFin in 48 hours, it won 1st place at Hack54’s Financial Hackathon.

🔗 Live Demo

View the live demo →

🧩 The Problem

MoMo agents are the backbone of everyday finance across Ghana, but they’re often left out of the financial system they help run. Managing liquidity — keeping enough cash and e-float on hand without money sitting idle — is a constant manual headache. And because most agents and their customers are unbanked, there’s very little structured data available for something as basic as credit scoring. MobiFin was built to close both gaps at once.

⚙️ How It Works

MobiFin is built around two connected intelligence layers:

Agent Intelligence — a real-time dashboard for cash and e-float tracking, demand forecasting, and rebalancing recommendations before liquidity problems happen (e.g. “rebalance ~GH₵4,000 before 10:30 AM”)
Alternative Credit Intelligence — with explicit customer consent, transactional behavior is turned into an explainable credit profile (alternative credit score, repayment probability, assessed risk) for customers who may lack a traditional credit history

Underneath both layers is a simple operational cycle: see the business (real-time transaction, cash, and e-float data), predict what’s coming (a forecasting model for demand and liquidity), act on it (rebalancing recommendations and alerts), and access opportunity (bridging agent and customer financial behavior to institutional partners).

💡 Technologies used

Python — custom machine learning model for liquidity forecasting and alternative credit scoring
JavaScript / React — agent and financial institution dashboards
Tailwind CSS — styling for the dashboard and demo experience
SQL — ledger, transaction, and credit profile data
Cloudflare — hosting and deployment
MTN API integration for USSD-based verification, pulling financial data directly from the mobile money network

🖼️ A Look Around the App

Agent onboarding:

Agent onboarding flow showing an account information form with full name, phone number, and password fields

Agent dashboard:

Agent dashboard showing a liquidity shortfall warning, cash on hand, e-float balance, commission earned, and business health score

Referring a customer for credit assessment:

Customer referrals view for an agent, showing a table of referred customers with consent status and lending decisions

Refer Customer for Loan modal with fields for customer identifier, phone number, requested loan amount, and lending institution

Consent via USSD:

Demo USSD simulation modal showing a simulated consent request awaiting a customer's approval

The financial institution’s side:

Forms Capital dashboard showing consented customers, credit-ready customers, average credit score, and indicative credit capacity

Credit underwriting portfolio view showing pending consent, consent granted, under review, and approved loan counts

🎯 What I Learned

The interesting engineering problem wasn’t the dashboard — it was building something useful for people the financial system usually skips over: agents managing float with no real-time visibility, and customers with real financial activity but no paper trail a bank recognizes. Forty-eight hours isn’t much time to train a model, wire up a live API integration, and have something demo-able that holds up under judges’ questions, so we scoped ruthlessly and kept coming back to whether an agent would actually want to use this, not just whether it worked in a demo.

🚀 Built by Team MobiFin at Hack54’s Financial Hackathon.