Skip to main content
Loading navigation
Back to challenges

CrewAI Office-to-Residential Structural Conversion Evaluator

Architect a multi-agent system with CrewAI and Cognition Devin that analyzes structural drawings and floor depths for office-to-residential conversions. The system evaluates MEP core placements and produces actionable conversion feasibility metrics with 90% confidence.

Business OperationsHosted by Vera
Sign in free to start

What you are building

Utilizes CrewAI multi-agent orchestration alongside Cognition Devin dynamic coding agents to process CAD metadata and structural office plans for residential adaptive reuse.

How work is evaluated

Evaluation

Evaluates CrewAI crew performance on structural feasibility analysis.

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

Structural Constraint Logic Check

Validates that high depth/post-tensioned slab triggers plumbing risk flags

Binary check

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

Dimension 2

Feasibility Score Variance

Variance against expert structural engineering feasibility baseline • target: 2 • range: 0-10

Binary check

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

Learning goals

  • Orchestrate collaborative CrewAI agents with dedicated architectural and MEP engineering roles

  • Integrate Cognition Devin workflows for automated CAD layout code parsing

  • Evaluate window perimeter ratios and floorplate core depth constraints

  • Generate cost-per-square-foot adaptive reuse scorecards

Resources

Dataset notes

Dataset of 30 commercial office floorplate plans with corresponding structural specs and residential conversion outcomes.

Challenge setup

Evaluates CrewAI crew performance on structural feasibility analysis.

Challenge input

JSON containing floor_depth_ft, core_to_window_distance_ft, slab_type, and column_grid_spacing

CrewAI

Orchestrates multi-role specialized engineering agents.

Cognition Devin

Executes CAD parser scripts in an automated sandbox.

Evaluated output

JSON with conversion_score (0-100), plumbing_chase_violation (bool), and structural_notes

Evaluation checks
  • Validates that high depth/post-tensioned slab triggers plumbing risk flags
  • Variance against expert structural engineering feasibility baseline • target: 2 • range: 0-10
Configured evaluation evidence
  • Benchmark: StructuralFeasibilityBench-CRE
  • Feasibility Score Variance target: 2
  • 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
  • Sacred · Optional

Evaluation contract

  • Structural Constraint Logic Check · Weight 1
  • Feasibility Score Variance · Weight 1

Metrics

  • Feasibility Score Variance · Target 2

Benchmarks

  • StructuralFeasibilityBench-CRE
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing floor_depth_ft, core_to_window_distance_ft, slab_type, and column_grid_spacing

    Output: JSON with conversion_score (0-100), plumbing_chase_violation (bool), and structural_notes

    Execution harness: Python

Evaluation tools

  • CrewAI · Agent Framework
  • Cognition Devin · Code Engine

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about CrewAI Office-to-Residential Structural Conversion Evaluator