Skip to main content
Back to challenges
Challenge

Build a MAS CPI Inflation Indexing and Fraud Anomaly Engine using Vercel AI SDK

Macroeconomic price fluctuations affect automated algorithmic loan underwriting parameters. Construct a Vercel AI SDK pipeline that integrates live MAS CPI data into credit risk evaluation models to maintain calibrated default predictions during inflation spikes.

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.

Develop a streaming TypeScript application using Vercel AI SDK to calibrate loan risk metrics against live MAS inflation indices.

Delivery guide

How work is evaluated

Evaluation

Evaluates loan default calibration accuracy when subjected to sudden CPI inflation shocks.

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

Inflation Stress Threshold Test

DSR > 50% under inflation adjustment must trigger REJECTED decision.

Binary check

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

Dimension 2

Risk Calibration Precision

Accuracy of adjusted default probabilities. • target: 0.94 • range: 0.88-1

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

  • Integrate live Monetary Authority of Singapore (MAS) CPI API feeds into Vercel AI SDK tools.

  • Dynamically adjust consumer disposable income estimates based on core vs headline inflation.

  • Detect borrowers exhibiting inflation-induced debt stress before defaults occur.

  • Stream calibrated loan underwriting decisions to front-end interfaces.

Resources and assets

Reference links and supporting material

Dataset notes

Synthetic Singapore retail borrower dataset paired with MAS monthly CPI series.

Challenge setup

Evaluates loan default calibration accuracy when subjected to sudden CPI inflation shocks.

Challenge input

JSON containing gross_monthly_income, existing_debt_commitments, mas_cpi_core_yoy_pct

AI SDK

Provides typed tool execution for external statistical API integration.

Evaluated output

JSON with adjusted_dsr_pct, underwriting_decision ('APPROVED', 'REJECTED'), risk_tier

Evaluation checks
  • DSR > 50% under inflation adjustment must trigger REJECTED decision.
  • Accuracy of adjusted default probabilities. • target: 0.94 • range: 0.88-1
Configured evaluation evidence
  • Benchmark: MacroCreditBench
  • Risk Calibration Precision target: 0.94
  • 1 public reference case
Configured runtime
  • JavaScript execution harness
  • javascript 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
  • Vercel AI SDK · Required
Orchestration
  • Vercel AI SDK · Required

Evaluation contract

  • Inflation Stress Threshold Test · Weight 1
  • Risk Calibration Precision · Weight 1

Metrics

  • Risk Calibration Precision · Target 0.94

Benchmarks

  • MacroCreditBench
  • security

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing gross_monthly_income, existing_debt_commitments, mas_cpi_core_yoy_pct

    Output: JSON with adjusted_dsr_pct, underwriting_decision ('APPROVED', 'REJECTED'), risk_tier

    Execution harness: JavaScript

Evaluation tools

  • AI SDK · Streaming & Tool Engine

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build a MAS CPI Inflation Indexing and Fraud Anomaly Engine using Vercel AI SDK