Back to evaluations
Draft evaluation

Multi-Agent Real Estate Underwriting Engine with AutoGen and Galileo — evaluation

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

Evaluation type
task based
Challenge
Multi-Agent Real Estate Underwriting Engine with AutoGen and Galileo
Difficulty
Advanced
Rigor
Not declared

The author has not specified a rigor level.

Evaluation overview

How the linked challenge is judged: tasks, benchmarks, and criteria count.

Tasks
1
Benchmarks
0
Criteria
0

Task templates

Inputs and expected outputs.

Task 1

Underwriting Execution Test

Runs the multi-agent pipeline on asset dossier JSONs and evaluates calculated metrics against target values.

Input format

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

Output format

JSON containing project_irr, target_dscr, net_equity_multiple, and risk_flags