B2B journey

Loan origination, end to end.

A lender's product team · personal loans · launching a digital funnel in one quarter

A lender wants a complete origination loop — discovery, KYC, score, decision, signature, audit — without stitching five vendors together.

Marketing claimcourtroom-replayable-audits
See the lender story Talk to us
The story · 5 scenes

Beginning to end.

A lender's product team · personal loans · launching a digital funnel in one quarter
·A lender wants a complete origination loop — discovery, KYC, score, decision, signature, audit — without stitching five vendors together.
  1. Scene 01 · Discover via AA
    An applicant consents to the AA share inside the lender's app. Mukodi receives the transaction history, categorises every line, and computes a financial health score before the form submits.
  2. Scene 02 · Verify the documents
    PAN and Aadhaar are submitted through the documents-KYC endpoint. Tamper flags, mask of the 8-digit Aadhaar number, and OCR confidence are returned in one bundle.
  3. Scene 03 · Score and explain
    Credit-risk fires with the SHAP attribution and a plain-language reason set. The Courtroom signs the verdict; the audit ledger captures the chain link.
  4. Scene 04 · Default-predict and band
    The lender requests the loan-default predictor on the candidate cohort. The output is a probability band by tenure, not a single point, so the pricing team can set tiers with their own policy.
  5. Scene 05 · Hand off and replay
    The decision is captured. Months later, the regulator visits; the lender replays the entire chain — discovery, KYC, score, default band, decision — from the audit ledger.
Helpers on call
The stack · helpers + models

Who's actually doing the work.

Features used
RBI Account AggregatorTransactions & insightsDocuments & KYCCredit risk + SHAP + counterfactualsHash-chained audit trailB2B API + Python / Java / Node SDKs
Honest limits

What this flow doesn't do.

  • The default-prediction band is a model output; the lender's pricing committee sets tiers under its licensed underwriting policy.
  • The AA discovery step requires the borrower's bank to be live on the AA network — most are; a few co-operative banks are still in pilot.
  • The audit chain stores hashes; lenders retain raw PII payloads in their own systems for replay under their data-retention regime.
← Back to all use cases