Skip to main content
Back to challenges
Challenge

Trade Credit Insurance Underwriting Agent for Credit Oman

Credit Oman approved over $589m in credit limits in H1 2026. Build an automated credit risk evaluation agent in Python to assess buyer credit limits, financial statements, and sovereign risk signals. Maintain underwriter scoring alignment within a 3% error margin relative to senior credit committee decisions.

Business OperationsHosted by Vera
Sign in free to start
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Build an AI-powered credit risk underwriting engine in Python to evaluate buyer creditworthiness and approve credit limits for trade credit insurance.

Delivery guide

How work is evaluated

Evaluation

Evaluates trade credit risk scoring and limit recommendation accuracy.

Datasets

Challenge datasets

Public datasets, plus the private datasets you have access to.

Evaluation rubric

How submissions are scored

These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

leverage_cap_test

Ensures companies with Debt/EBITDA > 4.5 receive restricted credit limits

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Dimension 2

limit_recommendation_accuracy

Alignment of approved limit against expert underwriting benchmark • target: 0.94 • range: 0-1

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

What you should walk away with

  • Parse financial statement metrics (Debt/EBITDA, Liquidity ratio, Working capital).

  • Integrate sovereign rating indicators and cross-border payment risk factors.

  • Determine recommended maximum credit line limits dynamically.

  • Generate structured underwriting rationale summaries for credit officers.

Resources and assets

Reference links and supporting material

Dataset notes

Anonymized trade financial profiles and credit approval outcomes from GCC commercial transactions.

Challenge setup

Evaluates trade credit risk scoring and limit recommendation accuracy.

Challenge input

JSON containing annual revenue, debt_to_ebitda, current_ratio, and country_risk_tier

Python Finance Tools

Calculates financial ratios and risk thresholds.

Evaluated output

JSON with credit_rating, approved_limit_usd, and recommendation

Evaluation checks
  • Ensures companies with Debt/EBITDA > 4.5 receive restricted credit limits
  • Alignment of approved limit against expert underwriting benchmark • target: 0.94 • range: 0-1
Configured evaluation evidence
  • Benchmark: CreditRiskBench
  • Limit Recommendation Accuracy target: 0.94
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python sandbox (unavailable on Versalist)

Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.

View technical recipe

Configured tools

Action Space
  • Writer · Required
  • MLflow · Optional
  • All Hands AI · Optional

Evaluation contract

  • leverage_cap_test · Weight 1
  • limit_recommendation_accuracy · Weight 1

Metrics

  • limit_recommendation_accuracy · Target 0.94

Benchmarks

  • CreditRiskBench
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing annual revenue, debt_to_ebitda, current_ratio, and country_risk_tier

    Output: JSON with credit_rating, approved_limit_usd, and recommendation

    Execution harness: Python

Evaluation tools

  • Python Finance Tools · Calculation Engine

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Trade Credit Insurance Underwriting Agent for Credit Oman