Skip to main content
Loading navigation
Back to challenges

Develop an Industrial Smelter Equipment Restoration Agent with Pydantic AI and Writer

Smelter restoration projects experience cost overruns due to unstructured site inspection reports and non-standardized replacement part specs. Build a structured Pydantic AI agent integrated with Writer to parse thermal camera telemetry and field logs, producing schema-validated work orders that reduce procurement clearance lead times by 30%.

Business OperationsHosted by Vera
Sign in free to start

What you are building

Build a type-safe inspection log parser and work order generator using Pydantic AI and Writer models for industrial aluminum smelter restoration.

How work is evaluated

Evaluation

Evaluates Pydantic schema validation adherence and extraction precision on smelter logs.

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

pydantic_schema_strictness

Output must strictly validate against Pydantic model without missing required fields

Binary check

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

Dimension 2

extraction_precision

Precision of extracted part numbers and defect categories • target: 0.95 • 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

  • Construct Pydantic validation models for industrial potline repair specifications

  • Utilize Pydantic AI agent classes to enforce strict type guarantees on LLM extractions

  • Connect Writer's Palms model for domain-specific industrial document understanding

  • Automate generation of SAP-compatible work orders with verified part numbers

Resources

Dataset notes

Unstructured potline inspection reports and thermal imaging logs from Al-Taweelah aluminium smelter.

Challenge setup

Evaluates Pydantic schema validation adherence and extraction precision on smelter logs.

Challenge input

Raw text log string from refractory inspection team.

Pydantic AI

Required framework for type-safe validation

Writer

Required domain-specific LLM platform

Evaluated output

JSON validated against WorkOrder Pydantic schema.

Evaluation checks
  • Output must strictly validate against Pydantic model without missing required fields
  • Precision of extracted part numbers and defect categories • target: 0.95 • range: 0-1
Configured evaluation evidence
  • Benchmark: Structured Industrial Extraction Benchmark
  • Extraction Precision target: 0.95
  • 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
  • Writer · Required
  • Pydantic AI · Optional
  • Zed · Optional
Orchestration
  • Pydantic AI · Optional

Evaluation contract

  • pydantic_schema_strictness · Weight 1
  • extraction_precision · Weight 1

Metrics

  • extraction_precision · Target 0.95

Benchmarks

  • Structured Industrial Extraction Benchmark
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: Raw text log string from refractory inspection team.

    Output: JSON validated against WorkOrder Pydantic schema.

    Execution harness: Python

Evaluation tools

  • Pydantic AI · Agent Framework
  • Writer · Model Runtime

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Develop an Industrial Smelter Equipment Restoration Agent with Pydantic AI and Writer