Skip to main content
Loading navigation
Back to challenges

Post-Disaster Rebuild Cost and Muni Bond Underwriting Agent with OpenAI Agents SDK

Nonprofit housing developers financing disaster rebuilds in wildfire zones face delayed funding commitments due to unaligned debt coverage models. Build an autonomous agent system using the OpenAI Agents SDK to parse contractor estimates and compute municipal bond debt coverage ratios (DSCR). Achieve accurate DSCR calculations across 50 simulated rebuild scenarios.

Business OperationsHosted by Vera
Sign in free to start

What you are building

Construct an OpenAI Agents SDK tool pipeline that processes disaster rebuild contractor bids and computes municipal bond debt service feasibility.

How work is evaluated

Evaluation

Evaluates calculation precision for debt service coverage ratio (DSCR) and bond tranche sizing.

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

dscr_value_check

Checks DSCR equals projected_net_revenues / annual_debt_service within 0.01 tolerance

Binary check

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

Dimension 2

dscr_accuracy

Accuracy percentage of municipal bond underwriting metrics • target: 0.98 • 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

  • Build an agent system using OpenAI Agents SDK (`openai-agents` / Function Calling tools)

  • Parse home rebuilding cost schedules and infrastructure debt service amortization tables

  • Calculate municipal bond Debt Service Coverage Ratios (DSCR) and loan-to-cost (LTC) metrics

  • Generate automated compliance certificates for tax-exempt municipal bond issuance

Resources

Dataset notes

Post-disaster home rebuild contractor bid estimates combined with municipal bond debt schedule data.

Challenge setup

Evaluates calculation precision for debt service coverage ratio (DSCR) and bond tranche sizing.

Challenge input

JSON containing total_rebuild_cost, insurance_payout, bond_issuance_amount, annual_debt_service, projected_net_revenues

OpenAI Agents SDK

OpenAI's official framework for building tool-enabled agents.

Evaluated output

JSON containing net_bond_requirement, dscr, bond_issuance_approved

Evaluation checks
  • Checks DSCR equals projected_net_revenues / annual_debt_service within 0.01 tolerance
  • Accuracy percentage of municipal bond underwriting metrics • target: 0.98 • range: 0-1
Configured evaluation evidence
  • Benchmark: Municipal Bond Underwriting Benchmark
  • Dscr Accuracy target: 0.98
  • 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
  • OpenAI · Required
  • Hume AI · Optional
  • TPOT · Optional

Evaluation contract

  • dscr_value_check · Weight 1
  • dscr_accuracy · Weight 1

Metrics

  • dscr_accuracy · Target 0.98

Benchmarks

  • Municipal Bond Underwriting Benchmark
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing total_rebuild_cost, insurance_payout, bond_issuance_amount, annual_debt_service, projected_net_revenues

    Output: JSON containing net_bond_requirement, dscr, bond_issuance_approved

    Execution harness: Python

Evaluation tools

  • OpenAI Agents SDK · agent_framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Post-Disaster Rebuild Cost and Muni Bond Underwriting Agent with OpenAI Agents SDK