# Personal loan terms in 40 seconds, with an underwriter on borderline cases

> A non-bank lender with 40,000 applications a year: scoring on its own loans, personal terms in 40 seconds instead of 1–2 days, gray-zone cases to a human.

URL: https://emilus.dev/en/cases/lending-pipeline/
Emilus · hello@emilus.dev · https://t.me/brrr28

1. [Emilus](https://emilus.dev/en/)
2. [Case studies](https://emilus.dev/en/cases/)
3. Lending pipeline

Custom software

emilus. Integrator

Status: In production

An underwriter used to check every application by hand, which took 1–2 days, and everyone got the same cautious rate. Now scoring built on the lender's own history returns personal terms in 40 seconds, and the underwriter receives gray-zone applications with the analysis already done.

At a glance

**Client:**
A non-bank lender in Ukraine≈40,000 applications a year

**Service:**
Custom software · emilus. Integrator · scoring and a lending pipeline

**Status:**
In production at the clientevery application runs through the pipeline

**By the numbers:**
Decision: 1–2 days → 40 sdefault rate −1.8 pp

The “before” numbers, the result and the payback are per client data.

## Where the client started

The lender receives about 40,000 applications a year, and an underwriter checked each one by hand. A decision took 1–2 days, and some approved clients borrowed elsewhere in the meantime.

Everyone paid the same rate, so it had to be cautious: too high for reliable clients, and still no filter for the ones who wouldn't pay back.

**≈40k**

applications a year

*per client data*

**1–2 days**

from application to decision

*per client data*

**1 rate**

the same for every client

*per client data*

The main rule The model calculates, and a person decides the borderline cases. Gray-zone applications go to an underwriter with the analysis ready, and the underwriter makes the call.

## What the system does

- **Takes the application on the website and collects everything up front:** amount, term, the client's details and documents, so the underwriter never has to chase anything.
- **Checks documents and identity automatically,** so a stolen or forged application never reaches a decision.
- **Scores risk with a model trained on the lender's own loans** and how they turned out, so decisions rest on how this lender's clients actually behave.
- **Calculates personal terms: rate, limit and schedule** for each client's risk, so a reliable client gets a better rate and a risky one a lower limit.
- **Shows the client ready terms in 40 seconds** with a payment schedule and a button to sign, so they don't go to another lender while they wait.
- **Decides clear-cut applications on its own,** so the underwriter doesn't spend time on the obvious.
- **Sends gray-zone applications to the underwriter with the analysis ready:** what speaks for the loan, what speaks against it, and which terms are possible. The underwriter makes the decision.
- **Logs every decision by the model and by people, with the reason,** so any loan can be explained and audited.
- **Retrains every month on loans whose repayment is already known,** so scoring keeps up with how clients behave now.
- **Puts a new model version to work only if it tests no worse than the previous one,** so decision quality can't slip unnoticed.

**Illustration** · How the system works

### The architecture in plain terms

An application from the website goes to verification with its documents, the model scores the risk on the lender's loan history, and the system calculates personal terms. Clear-cut applicants see their terms in 40 seconds; borderline ones go to the underwriter with an analysis.

**Data:**
applications and documents · issued loans and their repayment · credit history · underwriter decisions

**Roles:**
underwriter: the gray zone · client: terms and signing · risk manager: new model versions

## How it works

**From application to payout**

1. client · website · **Application** · Amount, term, documents
2. system · **Verification** · Documents and identity
3. model · **Scoring** · On the lender's own loan history
4. system · **Terms** · Rate, limit and schedule to match the risk
5. system · **40 seconds** · Clear-cut applications: terms go to the client at once
6. underwriter · **Gray zone** · A decision with the analysis ready
7. client · **Payout** · Signing and the money
8. model · **Retraining** · Monthly; only a version no worse than the last goes live

Defaults don't show up in the first week; they are counted on loans that have been running for several months. So −1.8 percentage points is the measured difference between loans issued after launch and loans issued before it, compared at the same age. It is not a model forecast.

## Results and payback

**40 s**

was 1–2 days

from application to personal terms on screen

per client data

**−1.8 pp**

portfolio default rate: fewer loans go unpaid

per client data

**+22%**

approved applications that end in a payout: clients no longer leave to wait elsewhere

per client data

**Project payback: 6–9 months *per client data*** · The gain here is in portfolio quality. The difference in defaults shows once loans issued under the new model have been through several months of repayment, so payback follows the credit cycle: 6–9 months.

## What's live

1. live Website applications with documents and automatic identity checks
2. live Scoring on the lender's own loans, with a personal rate, limit and schedule in 40 seconds
3. live Gray-zone cases go to the underwriter with the analysis ready, and every decision is logged
4. live Monthly model retraining, checked against the previous version

The service behind this case

## [Custom software](https://emilus.dev/en/services/custom-software/)

Systems built around the company's processes: scoring, client portals, CRM. The connections to document checks and credit history are [emilus. Integrator](https://emilus.dev/en/services/integrator/). Price after the brief.

## More case studies

- [A custom CRM and an AI sales agent for an electronics chain](https://emilus.dev/en/cases/electronics-store-chain/)
- [A booking platform for psychologists](https://emilus.dev/en/cases/b2b-saas-therapy-booking/)
- [Four AI agents for a car dealer hub](https://emilus.dev/en/cases/car-dealer-hub/)

[All case studies](https://emilus.dev/en/cases/)

Next step

## Write to us. A human replies.

[Message us on Telegram](https://t.me/brrr28)

[All contacts](https://emilus.dev/en/contact/)