Skip to main content
Back to challenges
Challenge

Google ADK & Cursor Reinsurance Exposure Concentration Analyzer

Global reinsurers like Munich Re and Lloyd's manage massive, highly dynamic risk books across global regions. Develop a risk aggregation and concentration analytics engine using Google ADK with Gemini models, built using Cursor workflow automation, to detect hyper-concentrated catastrophe treaty exposures.

Machine LearningHosted by Vera
Sign in free to start
Challenge brief

What you are building

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

Build a global reinsurance exposure monitor using Google ADK and Cursor to detect geographic risk concentrations.

Delivery guide

How work is evaluated

Evaluation

Evaluates geographic accumulation calculations and trigger alert precision.

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

math_correctness

Validates sum of TIV matches expected aggregate

Binary check

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

Dimension 2

h3_index_precision

Accuracy of spatial binning • target: 1 • range: 0-1

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

  • Build a Google ADK agent for processing treaty reinsurance bordereaux data

  • Leverage Cursor AI rules to accelerate complex spatial aggregation code generation

  • Compute spatial H3 grid exposure metrics across multi-region portfolios

  • Generate automated PML (Probable Maximum Loss) alert reports

Resources and assets

Reference links and supporting material

Dataset notes

Synthetic global reinsurance bordereaux dataset with insured property locations and values.

Challenge setup

Evaluates geographic accumulation calculations and trigger alert precision.

Challenge input

JSON list of property risk locations and insured values.

Google ADK

Agent workflows with Gemini integration

Cursor

Rapid agent tool development and rule enforcement

Evaluated output

JSON containing total aggregate risk, peak H3 cell, and threshold alert boolean.

Evaluation checks
  • Validates sum of TIV matches expected aggregate
  • Accuracy of spatial binning • target: 1 • range: 0-1
Configured evaluation evidence
  • Benchmark: ReinsuranceRiskBench
  • H3 Index Precision target: 1
  • 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 · Required
  • Cursor · Optional
  • Google ADK · Optional
Observation
  • Cursor · Optional

Evaluation contract

  • math_correctness · Weight 1
  • h3_index_precision · Weight 1

Metrics

  • h3_index_precision · Target 1

Benchmarks

  • ReinsuranceRiskBench
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON list of property risk locations and insured values.

    Output: JSON containing total aggregate risk, peak H3 cell, and threshold alert boolean.

    Execution harness: Python

Evaluation tools

  • Google ADK · Agent Platform
  • Cursor · Developer AI Environment

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Google ADK & Cursor Reinsurance Exposure Concentration Analyzer