Helpers on call
Retail journey
Income without a payslip.
Karthik, 31 · Swiggy partner + side gig delivery · Bengaluru
Karthik has earnings from three platforms and no payslip to show a lender. Mukodi turns the receipts into a story a credit team can read.
Marketing claimb2b-credit-risk-shap
The story · 4 scenes
Beginning to end.
Karthik, 31 · Swiggy partner + side gig delivery · Bengaluru
·Karthik has earnings from three platforms and no payslip to show a lender. Mukodi turns the receipts into a story a credit team can read.- Scene 01 · Add the earnings, any formatKarthik adds three months of payout PDFs, a few CSV exports from his delivery app, and a screenshot of his fuel-card statement. Mukodi reads all of them and writes one canonical ledger.
- Scene 02 · Verify the source documentsHis PAN and Aadhaar are OCR'd, cross-checked, and tamper-flagged in the app. The redacted artefacts stay in his account; the originals never leave his phone.
- Scene 03 · Get a credit-risk readMukodi computes a credit-risk score on his alternative-data profile — payout cadence, fuel spend, repayment history. Every factor is SHAP-explained and saved alongside the score.
- Scene 04 · Share with a partner lenderKarthik opens his lender's app and consents to a one-time share of his profile bundle. The lender sees the same Courtroom-signed bundle the app showed him.
Helpers on call
The stack · helpers + models
Who's actually doing the work.
Models invoked
Honest limits
What this flow doesn't do.
- Credit-risk scores are decisioning inputs for partner lenders — Mukodi does not approve, decline, or price loans itself.
- Document forgery flags are advisory; lenders may still require a human verification pass before they extend credit.
- Alternative-data scoring is gated to partner-lender flows; consumer-side scores ship as informational, not as a credit bureau replacement.