Skip to main content
Back to challenges
Challenge

Singapore Land Parcel Tender Yield Multi-Agent System with CrewAI

Urban land releases by authorities like Singapore's Urban Redevelopment Authority (URA) require rapid financial underwriting and plot ratio optimization. Build a multi-agent specialized CrewAI task force (Urban Planner, Financial Analyst, Bid Strategist) to evaluate site releases at Marina Gardens Lane and Orchard Boulevard. Achieve 90% alignment with optimal land tender bids.

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.

Deploy a CrewAI crew of specialized real estate agents to analyze Singapore URA site releases and compute optimal bid submission prices.

Delivery guide

How work is evaluated

Evaluation

Evaluates accuracy of GFA calculations and land bid price per square foot per plot ratio ($psf ppr).

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

gfa_accuracy

Verifies SQM to SQFT GFA calculation exactness.

Binary check

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

Dimension 2

bid_price_mae

Mean Absolute Error on recommended bid price psf ppr. • target: 25 • range: 0-200

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

  • Define specialized CrewAI Agents with custom goals, backstories, and tools

  • Set up sequential and hierarchical CrewAI process workflows for site evaluation

  • Parse URA site specifications including Gross Plot Ratio (GPR) and maximum Gross Floor Area (GFA)

  • Determine maximum viable land bid price based on target developer margin

Resources and assets

Reference links and supporting material

Dataset notes

Dataset of Singapore URA Government Land Sales (GLS) site specifications, plot ratios, and past winning tender bids.

Challenge setup

Evaluates accuracy of GFA calculations and land bid price per square foot per plot ratio ($psf ppr).

Challenge input

JSON object with site_area_sqm, gross_plot_ratio, expected_launch_price_psf, construction_cost_psf

CrewAI

Mandatory assigned multi-agent framework.

Evaluated output

JSON object with max_gfa_sqft, recommended_bid_psf_ppr

Evaluation checks
  • Verifies SQM to SQFT GFA calculation exactness.
  • Mean Absolute Error on recommended bid price psf ppr. • target: 25 • range: 0-200
Configured evaluation evidence
  • Benchmark: UrbanLandValuationBench
  • Bid Price Mae target: 25
  • 1 public reference case
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
  • crewAI · Optional
  • Zed · Optional

Evaluation contract

  • gfa_accuracy · Weight 1
  • bid_price_mae · Weight 1

Metrics

  • bid_price_mae · Target 25

Benchmarks

  • UrbanLandValuationBench
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON object with site_area_sqm, gross_plot_ratio, expected_launch_price_psf, construction_cost_psf

    Output: JSON object with max_gfa_sqft, recommended_bid_psf_ppr

    Execution harness: Python

Evaluation tools

  • CrewAI · Multi-agent Orchestrator

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Singapore Land Parcel Tender Yield Multi-Agent System with CrewAI