AI Credit Officer

Borrower Input

Fill borrower details and click Score Borrower.

Borrower ID

Monthly Income

Monthly Expense

Age

Has Smartphone?

Has Wallet?

Avg Wallet Balance

On-time Payment Ratio (0–1)

Number of Loans Taken

Result

Waiting for result...

Waiting for fraud analysis...

SHapley Additive exPlanations (SHAP)

Relative contribution of each feature to the credit decision

Decreases credit risk Increases credit risk

What Affected the Credit Decision?

Decreases
risk

Increases risk

Score a borrower to see SHAP explanations.

Overall summary will appear here after scoring.

AI Credit Officer Chat

Ask why the decision happened (grounded on ExplainChain + SHAP)

Hi! Score a borrower first, then ask: “Why did I get this decision?”

Model Performance

Live updating

Drift score: –

Stable

Stable (0.00–0.20)

Monitor (0.21–0.50)

Drift (0.51–1.00)

AUC

–

POOR (0.50–0.60) · FAIR (0.61–0.70) · GOOD (0.71–1.00)

Gini

–

LOW (0.00–0.30) · MEDIUM (0.31–0.50) · HIGH (0.51–1.00)

#	Version	Time	AUC	Δ	Actions

Model Maintenance (Admin)

Upload labeled training CSV and retrain ML models. Latest model is also saved as a versioned file.

Upload Training CSV

Last AUC: –

Rows used: –

Last trained: –

Model file: –

Drift: –

Borrower Scoring

Borrower Input

Result

SHapley Additive exPlanations (SHAP)

Model Performance

Model Maintenance (Admin)