Skip to main content
Loading navigation
Back to challenges

CrewAI Specialized Industry Risk Alliance Underwriter

Commercial insurance underwriters must collaborate across niche domain specialties to quote multi-peril commercial lines policies. Design a CrewAI multi-agent crew with specialized agents for construction, workers comp, and environmental risks to generate unified underwriting quotes.

Machine LearningHosted by Vera
Sign in free to start

What you are building

Orchestrate CrewAI role-playing agents to analyze cross-industry exposures and generate compliant underwriting guidelines.

How work is evaluated

Evaluation

Evaluates multi-agent crew consensus speed and coverage inclusion completeness across specialized perils.

Datasets

Public datasets, plus the private datasets you have access to.

How submissions are scored

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

line_item_completeness

Ensures all required commercial risk lines are included in the quote

Binary check

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

Dimension 2

pricing_consistency

Consistency score of rating manual calculations across agents • target: 0.95 • 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

  • Define specialized Agent roles, goals, and tools in CrewAI for different commercial lines

  • Configure sequential and hierarchical Task processes across agents

  • Synthesize disparate domain evaluations into a single policy proposal

Resources

Dataset notes

Commercial multi-line applicant dataset covering restaurant chains, construction contractors, and environmental remediators.

Challenge setup

Evaluates multi-agent crew consensus speed and coverage inclusion completeness across specialized perils.

Challenge input

JSON containing commercial applicant profile with restaurant + delivery operations

CrewAI

Orchestrates specialized domain agents in multi-line underwriting tasks

Evaluated output

JSON containing total_annual_premium, line_item_premiums, and special_endorsements

Evaluation checks
  • Ensures all required commercial risk lines are included in the quote
  • Consistency score of rating manual calculations across agents • target: 0.95 • range: 0-1
Configured evaluation evidence
  • Benchmark: MultiLine-Quote-Bench
  • Pricing Consistency target: 0.95
  • 1 public reference case
Configured runtime
  • 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

Action Space
  • CrewAI · Required
  • crewAI · Optional
  • Zed · Optional

Evaluation contract

  • line_item_completeness · Weight 1
  • pricing_consistency · Weight 1

Metrics

  • pricing_consistency · Target 0.95

Benchmarks

  • MultiLine-Quote-Bench
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing commercial applicant profile with restaurant + delivery operations

    Output: JSON containing total_annual_premium, line_item_premiums, and special_endorsements

    Execution harness: Python

Evaluation tools

  • CrewAI · Multi-Agent Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about CrewAI Specialized Industry Risk Alliance Underwriter