Challenge

Automate Agritech Underwriting via Claude Agents SDK and Azure

Agritech platforms struggle to rapidly assess smallholder farmer creditworthiness due to fragmented crop yields and financial records. Build an automated underwriting agent using the Claude Agents SDK and Azure AI infrastructure to synthesize satellite imagery logs and Account Aggregator statements. Achieve a 90% accuracy rating on credit risk classifications.

Business OperationsHosted by Vera
Challenge brief

What you are building

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

Build an intelligent agritech underwriting system with the Claude Agents SDK hosted on Azure AI services to automate rural credit scoring.

Delivery guide

How work is evaluated

Evaluation

Evaluates underwriting risk classifications against ground-truth agricultural loan portfolios.

Datasets

Shared data for this challenge

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

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

regulatory_bounds_check

Ensures credit recommendation strictly stays within allowable limits

Binary check

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

Dimension 2

underwriting_accuracy

Accuracy in matching ground-truth risk tiers • target: 0.9 • 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

What you should walk away with

  • Utilize Claude Agents SDK extended thinking capabilities for multi-step financial risk assessment

  • Integrate Azure OpenAI and Document Intelligence services for document parsing

  • Process Account Aggregator financial statements combined with crop yield telemetry

  • Generate compliant underwriting summary reports adhering to RBI lending guidelines

Resources and assets

Reference links and supporting material

Dataset notes

Agricultural yield logs combined with synthetic Account Aggregator cash flow reports across Indian districts.

How this agent runs

Evaluates underwriting risk classifications against ground-truth agricultural loan portfolios.

Challenge input

JSON object with crop harvest history and AA cashflow metrics

Claude Agents SDK

Provides multi-step reasoning capabilities for financial document processing.

Azure

Hosts enterprise models and enterprise security environments.

Evaluated output

JSON object with risk tier, max credit limit, and explanation

Checks for
  • Ensures credit recommendation strictly stays within allowable limits
  • Accuracy in matching ground-truth risk tiers • target: 0.9 • range: 0-1
Proof of success
  • Benchmark: Agritech Risk Scoring Benchmark
  • Underwriting Accuracy target: 0.9
  • 1 public reference case
Runtime evidence
  • Python execution harness
  • Python sandbox (unavailable on Versalist)
View technical recipe

Configured tools

Action Space
  • Azure · Required
  • Mindset AI · Optional
  • Retell AI · Optional

Evaluation contract

  • regulatory_bounds_check · Weight 1
  • underwriting_accuracy · Weight 1

Recipe state

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

Frequently Asked Questions about Automate Agritech Underwriting via Claude Agents SDK and Azure