Challenge

Multi-Agent Real Estate Underwriting Engine with AutoGen and Galileo

Commercial real estate firms face delayed underwriting decisions due to fragmented market metrics and valuation reports. Build a multi-agent system using AutoGen and Galileo to parse zoning, debt terms, and asset cash flows into automated investment memos. Achieve target underwriting memo generation with under 5% variance from expert benchmarks.

Business OperationsHosted by Vera
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Construct an AutoGen multi-agent pipeline integrated with Galileo observability to evaluate commercial acquisition targets and generate quantitative deal memos.

Delivery guide

How work is evaluated

Evaluation

Evaluates the precision of debt metrics calculation and memo completeness against benchmark deal sheets.

Datasets

Shared data for this challenge

Review public datasets and any private uploads tied to your build.

Loading datasets...
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

dscr_calculation_check

Validates DSCR is calculated within +/- 2% accuracy

Binary check

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

Dimension 2

irr_variance

Percentage variance between predicted IRR and target IRR • target: 0.01 • range: 0-0.05

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

  • Orchestrate multi-agent conversations between market research, debt compliance, and cash flow modeling agents using AutoGen

  • Integrate Galileo evaluation metrics to continuously monitor prompt quality, hallucination rate, and execution correctness

  • Parse unstructured commercial real estate dossiers and extract key cap rates, NOIs, and IRR targets

  • Synthesize automated, audit-ready investment memos with structured financial outputs

Resources and assets

Reference links and supporting material

Dataset notes

Synthetic commercial real estate asset packages including balance sheets, lease roll reports, and debt term sheets.

How this agent runs

Evaluates the precision of debt metrics calculation and memo completeness against benchmark deal sheets.

Challenge input

JSON containing property financial statements, NOI history, and debt terms

AutoGen

Multi-agent conversation orchestration for real estate tasks.

Galileo

Tracking evaluation metrics and detecting hallucination in agent output.

Evaluated output

JSON containing project_irr, target_dscr, net_equity_multiple, and risk_flags

Checks for
  • Validates DSCR is calculated within +/- 2% accuracy
  • Percentage variance between predicted IRR and target IRR • target: 0.01 • range: 0-0.05
Proof of success
  • Benchmark: CRE Financial Underwriting Benchmark
  • Irr Variance target: 0.01
  • 1 public reference case
Runtime evidence
  • Python execution harness
  • Python sandbox
View technical recipe

Configured tools

Action Space
  • AutoGen · Optional
  • Fixie · Optional
Observation
  • Galileo · Required
Reward / Eval
  • Galileo · Required

Evaluation contract

  • dscr_calculation_check · Weight 1
  • irr_variance · Weight 1

Recipe state

This is a preview. The configuration can change before the evaluation recipe is locked.

Frequently Asked Questions about Multi-Agent Real Estate Underwriting Engine with AutoGen and Galileo