Build a Multi-Rail Merchant Fee Optimization Crew with CrewAI
Shifting debit fee regulations drive merchants to evaluate alternative routing options across regional debit networks. Build a multi-agent collaborative crew using CrewAI where specialized agents analyze fee schedules, route transactions, and audit settlement math.
What you are building
Orchestrate a team of specialized CrewAI agents (Fee Analyst, Routing Agent, Auditor) to optimize debit card network selection.
How work is evaluated
Evaluates cost savings and routing choice accuracy produced by the agent crew.
Datasets
Public datasets, plus the private datasets you have access to.
How submissions are scored
test_least_cost_selection
Verify cheapest network selected for non-exempt ticket
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
optimization_accuracy
Percentage of transactions routed to true minimum fee network • target: 0.98 • range: 0.9-1
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
Learning goals
Configure role-playing agents in CrewAI (Fee Auditor, Route Optimizer, Compliance Verifier).
Orchestrate hierarchical or sequential task delegation across agents.
Deliver automated debit processing routing maps with verified margin optimization.
Resources
Synthetic debit transaction batches with interchange rate structures.
Challenge setup
Evaluates cost savings and routing choice accuracy produced by the agent crew.
Challenge input
JSON transaction batch profile with ticket sizes and network options
CrewAI
Framework requirement
Evaluated output
JSON with recommended_network, total_fee, projected_savings
- Verify cheapest network selected for non-exempt ticket
- Percentage of transactions routed to true minimum fee network • target: 0.98 • range: 0.9-1
- Benchmark: Merchant Optimization Benchmark v1
- Optimization Accuracy target: 0.98
- 1 public reference case
- Python execution harness
- Python sandbox (unavailable on Versalist)
Shows the first task, up to three tools, and up to three evidence items.
View technical recipe
Configured tools
- CrewAI · Required
- crewAI · Optional
Evaluation contract
- test_least_cost_selection · Weight 1
- optimization_accuracy · Weight 1
Metrics
- optimization_accuracy · Target 0.98
Benchmarks
- Merchant Optimization Benchmark v1
- general
Reference cases
- 1 public reference case
Workflow tasks
- Task 1
Input: JSON transaction batch profile with ticket sizes and network options
Output: JSON with recommended_network, total_fee, projected_savings
Execution harness: Python
Evaluation tools
- CrewAI · agent_framework
Recipe freeze marker
No recipe freeze time is recorded.