Skip to main content
Loading navigation
Back to challenges

World Cup 2030 Stadium Procurement Negotiation with AutoGen

Morocco's $291m stadium package tender for World Cup 2030 requires multi-party negotiation between civil engineering contractors and FIFA inspectors. Build an AutoGen multi-agent system where Contractor, FIFA Auditor, and Procurement Manager agents interact conversationally to resolve bid compliance issues. Achieve agreement within 5 agent turns while satisfying all safety requirements.

Business OperationsHosted by Vera
Sign in free to start

What you are building

Build an AutoGen multi-agent conversational framework for World Cup 2030 stadium tender negotiations.

How work is evaluated

Evaluation

Evaluates AutoGen conversational convergence on World Cup stadium bid terms.

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

Capacity Minimum Check

Ensures agreed_capacity is at or above FIFA minimum requirement.

Binary check

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

Dimension 2

Negotiation Efficiency Score

Score based on reaching consensus in under 5 conversation turns. • 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

  • Configure multi-agent conversations using Microsoft AutoGen `ConversableAgent`

  • Define agent personas: Contractor, FIFA Compliance Officer, and Tender Manager

  • Implement group chat mechanisms with custom turn-taking logic

  • Automate generation of finalized procurement contracts

Resources

Dataset notes

Morocco World Cup 2030 tender requirements and FIFA stadium regulations.

Challenge setup

Evaluates AutoGen conversational convergence on World Cup stadium bid terms.

Challenge input

JSON object with initial_proposed_budget_usd, proposed_capacity, fifa_min_capacity

AutoGen

Assigned framework for multi-agent conversational negotiations.

Evaluated output

JSON object with consensus_reached, agreed_budget_usd, agreed_capacity, turn_count

Evaluation checks
  • Ensures agreed_capacity is at or above FIFA minimum requirement.
  • Score based on reaching consensus in under 5 conversation turns. • target: 1 • range: 0-1
Configured evaluation evidence
  • Benchmark: NegotiationBench
  • Negotiation Efficiency Score target: 1
  • 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
  • AutoGen · Required
  • Kubeflow · Optional
  • Bland AI · Optional

Evaluation contract

  • Capacity Minimum Check · Weight 1
  • Negotiation Efficiency Score · Weight 1

Metrics

  • Negotiation Efficiency Score · Target 1

Benchmarks

  • NegotiationBench
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON object with initial_proposed_budget_usd, proposed_capacity, fifa_min_capacity

    Output: JSON object with consensus_reached, agreed_budget_usd, agreed_capacity, turn_count

    Execution harness: Python

Evaluation tools

  • AutoGen · Multi-Agent Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about World Cup 2030 Stadium Procurement Negotiation with AutoGen