Credit for the credit-invisible.
Most MSMEs in India can't get a formal loan — not because they're not creditworthy, but because they lack the documents traditional underwriting demands. This is an AI/ML Financial Health Card that scores a business from the alternate data it already generates — GST, UPI, bank (AA), EPFO — and explains every point.
The problem
New-to-Credit and New-to-Bank MSMEs are rejected at high rates for missing financial documents, despite rich alternate data. The absence of a unified assessment framework misses viable borrowers and slows financial inclusion.
The solution
A transparent WOE/logistic scorecard (the way banks actually build credit scores) over seven health dimensions — including an Operational Footprint built from utility/power-consumption data for thin-file, new-to-credit businesses — validated by an XGBoost challenger. Every score decomposes into visible reason codes and weights — explainable by construction, not a black box.
Ecosystem-ready
Built around the rails: Account Aggregator (consent), Unified Lending Interface (data), OCEN (decisioning). The engine is data-source-agnostic — synthetic today, real feeds on integration, with no change to the scoring code.
Why us
Built by a solo full-stack developer with a finance background who owns an MSME — the exact credit-invisible business this serves. The problem is lived, not theorised.
Demo runs entirely on a synthetic generator whose schema mirrors the real rails — no customer data is used. See How it works for the data each dimension needs and the synthetic→real swap.