Skip to main content
Back to challenges
Challenge

Cross-Border Maritime Trade Discrepancy Resolver with CrewAI and Groq Cloud

Build an automated cross-border trade document review crew using CrewAI and Groq Cloud high-speed inference. Reconcile bills of lading, customs declarations, and port clearance records to accelerate vessel turnaround at Singapore ports.

Workflow AutomationHosted by Vera
Sign in free to start
Challenge brief

What you are building

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

Orchestrate multi-role CrewAI agents running on Groq Cloud ultra-fast inference to detect discrepancies across maritime trade documents.

Delivery guide

How work is evaluated

Evaluation

Evaluates identification of weight mismatch, HS code discrepancy, and cargo clearance recommendations.

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

discrepancy_detected

Ensures weight mismatch triggers clearance hold

Binary check

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

Dimension 2

execution_speed_ms

Time taken to complete full crew evaluation cycle • target: 1200 • range: 0-3000

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

  • Design collaborate multi-agent workflows using CrewAI Agent, Task, and Crew primitives

  • Configure Groq Cloud API connections for sub-second inference across document comparison tasks

  • Implement trade document validation logic matching HS Codes, weights, and consignee data

  • Generate automated clearance hold or permit approval notices for Singapore customs

Resources and assets

Reference links and supporting material

Dataset notes

Synthetic dataset of 30 paired shipping manifests and Singapore customs entry records.

Challenge setup

Evaluates identification of weight mismatch, HS code discrepancy, and cargo clearance recommendations.

Challenge input

JSON containing bill_of_lading and customs_declaration objects

CrewAI

Orchestrates dedicated agent roles for manifest and customs auditing.

Groq Cloud

Delivers sub-second response times for complex trade validation.

Evaluated output

JSON containing clearance_status, discrepancy_found (boolean), and discrepancy_details

Evaluation checks
  • Ensures weight mismatch triggers clearance hold
  • Time taken to complete full crew evaluation cycle • target: 1200 • range: 0-3000
Configured evaluation evidence
  • Benchmark: Logistics Efficiency Benchmark
  • Execution Speed Ms target: 1200
  • 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
  • CrewAI · Required
  • crewAI · Optional
Policy Serving
  • Groq Cloud · Optional

Evaluation contract

  • discrepancy_detected · Weight 1
  • execution_speed_ms · Weight 1

Metrics

  • execution_speed_ms · Target 1200

Benchmarks

  • Logistics Efficiency Benchmark
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing bill_of_lading and customs_declaration objects

    Output: JSON containing clearance_status, discrepancy_found (boolean), and discrepancy_details

    Execution harness: Python

Evaluation tools

  • CrewAI · Agent Roles Framework
  • Groq Cloud · LLM Acceleration

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Cross-Border Maritime Trade Discrepancy Resolver with CrewAI and Groq Cloud