Skip to main content
Back to challenges
Challenge

Build a Credit Card Default Risk & Early Warning Agent with LlamaIndex RAG

Macroeconomic pressures have driven credit card defaults to multi-year highs. Build a advanced RAG compliance agent using LlamaIndex to query complex borrower credit histories, macroeconomic indices, and loan documentation to forecast default probabilities with an AUC-ROC > 0.88.

CybersecurityHosted by Vera
Sign in free to start
Challenge brief

What you are building

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

Construct an enterprise credit risk modeling and early warning system using LlamaIndex RAG data connectors and advanced query engines.

Delivery guide

How work is evaluated

Evaluation

Evaluates default prediction AUC-ROC and retrieval context relevance for early warning triggers.

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

auc_roc_pass

Default prediction AUC-ROC must exceed 0.88 baseline.

Binary check

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

Dimension 2

auc_roc

Area under ROC curve for credit default identification. • target: 0.91 • range: 0.8-1

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

  • Implement advanced RAG indexing over credit agreements using LlamaIndex

  • Construct hybrid vector-keyword query engines with reranking layers

  • Predict credit card default probability based on structured balance and macro data

  • Generate early warning debt restructuring recommendations for high-risk accounts

Resources and assets

Reference links and supporting material

Dataset notes

Dataset of borrower credit profiles, historical repayment patterns, and unstructured debt counseling transcripts.

Challenge setup

Evaluates default prediction AUC-ROC and retrieval context relevance for early warning triggers.

Challenge input

JSON borrower profile detailing debt ratio, payment delay days, and income

LlamaIndex

Provides state-of-the-art query engines and data connectors for complex indexing.

Evaluated output

JSON default prediction probability and primary risk factor summary

Evaluation checks
  • Default prediction AUC-ROC must exceed 0.88 baseline.
  • Area under ROC curve for credit default identification. • target: 0.91 • range: 0.8-1
Configured evaluation evidence
  • Benchmark: Credit Risk ML Suite
  • Auc Roc target: 0.91
  • 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
  • Llama Index · Required
  • LlamaIndex · Optional
  • Llama 3.2 · Optional

Evaluation contract

  • auc_roc_pass · Weight 1
  • auc_roc · Weight 1

Metrics

  • auc_roc · Target 0.91

Benchmarks

  • Credit Risk ML Suite
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON borrower profile detailing debt ratio, payment delay days, and income

    Output: JSON default prediction probability and primary risk factor summary

    Execution harness: Python

Evaluation tools

  • LlamaIndex · RAG Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build a Credit Card Default Risk & Early Warning Agent with LlamaIndex RAG