Skip to main content
Back to challenges
Challenge

Casualty ILS Portfolio Risk Optimizer using Multi-Agent CrewAI

Structuring casualty insurance-linked securities requires balancing long-tail liability risks against capital return benchmarks. You will construct a multi-agent investment analysis team using CrewAI to evaluate casualty ILS tranches and optimize portfolio allocation under macroeconomic stress scenarios. Achieve a Sharpe ratio improvement of at least 18% over standard baseline strategies.

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.

Orchestrate specialized actuarial and financial agents using CrewAI to construct optimized casualty ILS portfolios.

Delivery guide

How work is evaluated

Evaluation

Assess portfolio Sharpe ratio uplift achieved by the multi-agent optimization crew.

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

weights_sum_to_one

Allocation weights must sum to 1.0 (+/- 0.01).

Binary check

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

Dimension 2

sharpe_improvement_pct

Relative percentage increase in portfolio Sharpe ratio over baseline • target: 0.18 • 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

  • Define specialized agent roles and goals in CrewAI for capital market research

  • Simulate casualty tail-risk scenarios using heavy-tailed probability functions

  • Build multi-agent task execution sequences for bond selection and risk budgeting

  • Maximize risk-adjusted portfolio returns measured by the Sharpe ratio

Resources and assets

Reference links and supporting material

Dataset notes

Synthetic dataset of casualty ILS bond yields, underlying liability distribution curves, and macro interest rate histories.

Challenge setup

Assess portfolio Sharpe ratio uplift achieved by the multi-agent optimization crew.

Challenge input

JSON array of candidate casualty ILS tranches with yield and tail loss metrics

CrewAI

Multi-agent task orchestration with distinct specialist roles.

Evaluated output

JSON allocation weights and expected Sharpe ratio

Evaluation checks
  • Allocation weights must sum to 1.0 (+/- 0.01).
  • Relative percentage increase in portfolio Sharpe ratio over baseline • target: 0.18 • range: 0-1
Configured evaluation evidence
  • Benchmark: ILSOptimBench-2026
  • Sharpe Improvement Pct target: 0.18
  • 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
  • Zhipu AI · Optional

Evaluation contract

  • weights_sum_to_one · Weight 1
  • sharpe_improvement_pct · Weight 1

Metrics

  • sharpe_improvement_pct · Target 0.18

Benchmarks

  • ILSOptimBench-2026
  • energy

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON array of candidate casualty ILS tranches with yield and tail loss metrics

    Output: JSON allocation weights and expected Sharpe ratio

    Execution harness: Python

Evaluation tools

  • CrewAI · Agent Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Casualty ILS Portfolio Risk Optimizer using Multi-Agent CrewAI