Financeml

Credit / default-risk scoring

Predict 12-month default probability from borrower features (income, history, debt-to-income, credit-bureau signals). Retail and SME underwriting workload.

Configure your run

Data scale

How big is the problem you're trying to solve?

Accuracy you need

What error tolerance can your downstream decision absorb?

Total project budget (USD)

We use this to filter candidate devices. We divide by the typical run count for this workload — you can override that under Advanced.

Advanced— response time, audit level, label

Response time

Audit level

'Regulated' triggers automatic decision-record + bilingual PDF + ZIP bundle.

Workload label (optional)

Free text — rides into the decision record so audits can grep your reference (e.g. "AAPL Dec 2026 $200 call").

The same circuit shape and ranking are produced by qlro.recommend_workload("industry.finance.credit_default_scoring") in the Python SDK — useful if you want to automate this in CI / CD.