Skip to main content
Loading navigation
Back to challenges

CrewAI & Daytona Multi-Agent Capital Allocation Engine

Major real estate developers restarting development engines need cross-departmental coordination across underwriting, risk, and design. Build a CrewAI multi-agent team inside isolated Daytona sandbox environments to evaluate mixed-use parcel development potential.

Business OperationsHosted by Vera
Sign in free to start

What you are building

Orchestrate role-playing CrewAI agents executed inside Daytona environments to perform end-to-end development underwriting.

How work is evaluated

Evaluation

Evaluates multi-agent investment memo synthesis and financial metric calculations.

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

irr_threshold_check

Ensures projected IRR matches mathematical discount rate models within 2%.

Binary check

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

Dimension 2

memo_completeness_score

Score measuring presence of market, engineering, and financial sub-sections. • 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

  • Create specialized CrewAI agents for Market Analyst, Structural Engineer, and Underwriter

  • Configure Daytona workspace environments for secure code and model execution

  • Implement collaborative agent tasks using CrewAI sequential and hierarchical workflows

  • Generate comprehensive real estate pro forma financial reports automatically

Resources

Dataset notes

100 commercial development parcel proposals including zoning codes, cost benchmarks, and market rent metrics.

Challenge setup

Evaluates multi-agent investment memo synthesis and financial metric calculations.

Challenge input

JSON parcel characteristics and construction cost assumptions

CrewAI

Provides multi-agent task delegation and role-playing capabilities.

Daytona

Provides isolated, standard environments to execute agent code and tools.

Evaluated output

JSON containing estimated_cost, projected_irr, recommendation

Evaluation checks
  • Ensures projected IRR matches mathematical discount rate models within 2%.
  • Score measuring presence of market, engineering, and financial sub-sections. • target: 0.95 • range: 0-1
Configured evaluation evidence
  • Benchmark: CREUnderwritingBench
  • Memo Completeness Score 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

Environment
  • Daytona · Optional
Action Space
  • CrewAI · Required
  • Daytona · Optional
  • Aim · Optional

Evaluation contract

  • irr_threshold_check · Weight 1
  • memo_completeness_score · Weight 1

Metrics

  • memo_completeness_score · Target 0.95

Benchmarks

  • CREUnderwritingBench
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON parcel characteristics and construction cost assumptions

    Output: JSON containing estimated_cost, projected_irr, recommendation

    Execution harness: Python

Evaluation tools

  • CrewAI · Agent Framework
  • Daytona · Development Sandbox

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about CrewAI & Daytona Multi-Agent Capital Allocation Engine