Skip to main content
Back to challenges
Challenge

Build an Adaptive Payment Routing and Fraud Minimization Agent with OpenAI Agents SDK

Dynamic payment routing must balance transaction fees, settlement latency, and authorization risk in real time. Implement an OpenAI Agents SDK system that routes multi-currency payments to optimal gateways while maintaining authorization rates above 99%.

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 an autonomous payment routing agent using OpenAI Agents SDK to select ideal payment acquirers dynamically.

Delivery guide

How work is evaluated

Evaluation

Simulates payment gateway traffic to evaluate authorization optimization and fee minimization.

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

Fee Minimization Test

Must select local debit network over standard card network for low-risk debit transactions.

Binary check

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

Dimension 2

Authorization Rate Success

Overall simulated authorization rate. • target: 0.995 • range: 0.98-1

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 multi-turn autonomous payment routing flows using OpenAI Agents SDK.

  • Define function calling schemas for real-time gateway health and fee checks.

  • Incorporate adaptive fraud probability heuristics before routing transactions.

  • Achieve cost savings while enforcing 99%+ transaction authorization rate.

Resources and assets

Reference links and supporting material

Dataset notes

Transaction log simulation dataset with real-time acquirer availability and fee matrices.

Challenge setup

Simulates payment gateway traffic to evaluate authorization optimization and fee minimization.

Challenge input

JSON containing card_type, amount, currency, merchant_category, and user_risk

OpenAI Agents SDK

Provides function calling for continuous tool invocation.

Evaluated output

JSON with chosen_gateway, estimated_fee, expected_authorization_pct

Evaluation checks
  • Must select local debit network over standard card network for low-risk debit transactions.
  • Overall simulated authorization rate. • target: 0.995 • range: 0.98-1
Configured evaluation evidence
  • Benchmark: PaymentRouteBench
  • Authorization Rate Success target: 99.5%
  • 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
  • OpenAI · Required
  • Azure OpenAI Service · Optional
  • OpenAI o3 · Optional

Evaluation contract

  • Fee Minimization Test · Weight 1
  • Authorization Rate Success · Weight 1

Metrics

  • Authorization Rate Success · Target 0.995

Benchmarks

  • PaymentRouteBench
  • security

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing card_type, amount, currency, merchant_category, and user_risk

    Output: JSON with chosen_gateway, estimated_fee, expected_authorization_pct

    Execution harness: Python

Evaluation tools

  • OpenAI Agents SDK · Agent Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build an Adaptive Payment Routing and Fraud Minimization Agent with OpenAI Agents SDK