Skip to main content
Back to challenges
Challenge

Orchestrate a CrewAI Industrial Feasibility Squad with Pydantic AI for GCC Green Iron Projects

Oman's Meranti green iron project and Saudi Arabia's push to localize desalination production require rigorous feasibility analysis. This challenge tasks you with building a multi-agent system using CrewAI to evaluate the economic and technical viability of these industrial projects. You will create three specialized agents: a 'Regional Compliance Expert', a 'Supply Chain Analyst', and a 'Sustainability Consultant'. Each agent will utilize Pydantic AI for structured data validation, ensuring that all findings (e.g., water requirements, carbon footprint, local content spend) adhere to strict data schemas. The agents must collaborate to produce a final 'Go/No-Go' report that accounts for Oman's water supply constraints from Marafiq and Saudi's localization targets. The goal is to move from manual PDF-based feasibility studies to an automated, structured agentic analysis pipeline.

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.

Build a collaborative multi-agent squad using CrewAI and Pydantic AI to automate industrial feasibility studies for GCC green energy and manufacturing projects.

Delivery guide

How work is evaluated

Evaluation

The system is evaluated on the structural integrity of its findings and its ability to synthesize data from disparate sources into a cohesive feasibility report.

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

Schema Validation

Validates that the output exactly matches the Pydantic FeasibilityModel.

Binary check

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

Dimension 2

Collaborative Synthesis

Quality score for how well agents integrated each other's data • target: 4.5 • range: 1-5

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 and orchestrate role-playing AI agents with CrewAI for complex industrial analysis

  • Implement strict data validation for agent outputs using Pydantic AI models

  • Develop collaborative task dependencies where agent outputs feed into subsequent processes

  • Analyze water-energy nexus constraints in GCC green hydrogen and iron projects

Resources and assets

Reference links and supporting material

Dataset notes

Project profiles for GCC industrial initiatives including desalination localized production specs, Marafiq water volume data, and Oman green iron technical requirements.

Challenge setup

The system is evaluated on the structural integrity of its findings and its ability to synthesize data from disparate sources into a cohesive feasibility report.

Challenge input

Project description and resource constraints

CrewAI

Best-in-class framework for multi-agent role-playing workflows.

Pydantic AI

Ensures AI agent outputs conform to industrial standards.

Evaluated output

Structured JSON following a Pydantic schema

Evaluation checks
  • Validates that the output exactly matches the Pydantic FeasibilityModel.
  • Quality score for how well agents integrated each other's data • target: 4.5 • range: 1-5
Configured evaluation evidence
  • Benchmark: Multi-Agent Logic Benchmark
  • Collaborative Synthesis target: 4.5
  • Protected reference cases support private evaluation
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
  • Pydantic AI · Optional
  • Zed · Optional
Orchestration
  • Pydantic AI · Optional

Evaluation contract

  • Schema Validation · Weight 1
  • Collaborative Synthesis · Weight 1

Metrics

  • Collaborative Synthesis · Target 4.5

Benchmarks

  • Multi-Agent Logic Benchmark
  • general

Reference cases

  • Protected reference cases are configured.

Workflow tasks

  1. Task 1

    Input: Project description and resource constraints

    Output: Structured JSON following a Pydantic schema

    Execution harness: Python

Evaluation tools

  • CrewAI · Agent Orchestration
  • Pydantic AI · Data Integrity

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Orchestrate a CrewAI Industrial Feasibility Squad with Pydantic AI for GCC Green Iron Projects