Skip to main content
Back to challenges
Challenge

Build a Seaborne Fuel Trade Tracking Agent with Google ADK

Tracking shifts in seaborne fuel oil imports requires automated tanker movement and refinery yield analysis. Build an autonomous tracking service using Google ADK and Gemini 1.5 Pro to monitor tanker shipments, parse port manifests, and compute regional import volumes.

Business OperationsHosted by Vera
Sign in free to start
Challenge brief

What you are building

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

Utilize Google ADK and Vertex AI to process vessel tracking signals and verify seaborne fuel oil trade flows into Saudi ports.

Delivery guide

How work is evaluated

Evaluation

Evaluates tanker tonnage extraction precision and monthly import aggregation accuracy.

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

tonnage_sum_check

Verify aggregated total tonnage matches sum of input manifests

Binary check

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

Dimension 2

aggregation_precision

Precision of manifest data extraction and math verification • target: 0.99 • range: 0.92-1

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 Google ADK agents for continuous maritime trade telemetry ingestion

  • Parse shipping manifests and bill of lading documents using Gemini vision capabilities

  • Calculate monthly fuel oil import volumes categorized by origin country

  • Output automated energy market intelligence alerts for trade compliance officers

Resources and assets

Reference links and supporting material

Dataset notes

Tanker AIS discharge events, bill of lading document extracts, and monthly Saudi fuel oil import statistics.

Challenge setup

Evaluates tanker tonnage extraction precision and monthly import aggregation accuracy.

Challenge input

JSON manifest batch

Google ADK

Google ADK provides native integration with Gemini models for vision/document extraction

Evaluated output

JSON aggregated import summary

Evaluation checks
  • Verify aggregated total tonnage matches sum of input manifests
  • Precision of manifest data extraction and math verification • target: 0.99 • range: 0.92-1
Configured evaluation evidence
  • Benchmark: CommodityTradeBench
  • Aggregation Precision target: 0.99
  • 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
  • Google ADK · Required
  • Google · Optional
  • Google Gemini · Optional

Evaluation contract

  • tonnage_sum_check · Weight 1
  • aggregation_precision · Weight 1

Metrics

  • aggregation_precision · Target 0.99

Benchmarks

  • CommodityTradeBench
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON manifest batch

    Output: JSON aggregated import summary

    Execution harness: Python

Evaluation tools

  • Google ADK · Agent Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build a Seaborne Fuel Trade Tracking Agent with Google ADK