Skip to main content
Back to challenges
Challenge

Deploy Type-Safe Multi-Agent Orchestration with Pydantic AI and Claude Sonnet 4.6.6

Create a robust agent team for corporate policy analysis and regulatory compliance monitoring, inspired by SAP's recent antitrust regulatory pivot. This challenge focuses on building type-safe agent workflows where every action and response is validated through Pydantic models. You will coordinate between specialized agents using Mastra AI to ensure compliant and structured communication between system components.

Special Purpose AgentsHosted by Vera
Sign in free to start
Challenge brief

What you are building

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

Implement a type-safe multi-agent team for policy monitoring using Pydantic AI and Claude Sonnet 4.6.6.

Delivery guide

How work is evaluated

Evaluation

Evaluate agent compliance with policy definitions.

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

SchemaValidation

Ensure output matches pydantic model

Binary check

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

Dimension 2

ComplianceRate

Percentage of successful validations • target: 1 • 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

  • Master Pydantic AI dependency injection and schema enforcement for LLM outputs

  • Orchestrate complex agent team workflows using the Mastra AI framework

  • Implement tracing and evaluation loops using Arize Phoenix for observability

  • Design structured communication protocols between specialized agents

  • Leverage Cursor for AI-assisted development of compliance checking modules

  • Utilize Bito AI for rapid generation of edge-case scenarios to test agent policy adherence

Resources and assets

Reference links and supporting material

Dataset notes

Sample data for 1 tasks

Challenge setup

Evaluate agent compliance with policy definitions.

Challenge input

Policy text

Pydantic AI

Typed Python agent framework.

Mastra AI

TypeScript agent framework.

Bito AI

AI assistant for dev tasks

Evaluated output

Validated JSON

Evaluation checks
  • Ensure output matches pydantic model
  • Percentage of successful validations • target: 1 • range: 0-1
Configured evaluation evidence
  • ComplianceRate target: 100%
  • 1 public reference case
Configured runtime
  • Python execution harness

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
  • Pydantic AI · Required
  • Mastra AI · Optional
  • Bito AI · Optional
Orchestration
  • Pydantic AI · Required
  • Mastra AI · Optional

Evaluation contract

  • SchemaValidation · Weight 1
  • ComplianceRate · Weight 1

Metrics

  • ComplianceRate · Target 1

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: Policy text

    Output: Validated JSON

    Execution harness: Python

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Deploy Type-Safe Multi-Agent Orchestration with Pydantic AI and Claude Sonnet 4.6.6