Back to challenges
Challenge

Industrial Equipment Vibration Diagnostics Agent using Claude Agents SDK

Unplanned mechanical breakdown of rotating high-speed packaging equipment causes severe line downtime. Build a diagnostic agent using Anthropic's Claude Agents SDK with extended thinking enabled to analyze high-frequency vibration spectrums and thermography logs. Identify root-cause bearing degradation modes with a 93% diagnostic accuracy rate.

EngineeringHosted 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 an equipment diagnostic agent using Claude Agents SDK with extended thinking to reason over multi-sensor vibration FFT spectrums.

Delivery guide

How work is evaluated

Evaluation

Evaluates accuracy of bearing fault mode detection from spectral peak frequencies.

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

valid_fault_type

Ensure fault_mode matches standard bearing defect classifications

Binary check

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

Dimension 2

diagnostic_accuracy

Ratio of correct fault mode diagnoses across evaluation suite • target: 0.93 • 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 with Extended Thinking parameters for step-by-step diagnostic reasoning

  • Analyze fast Fourier transform (FFT) peak distributions alongside bearing geometry specs

  • Output precise fault classification alerts detailing failure stage and recommended maintenance interval

Resources and assets

Reference links and supporting material

Dataset notes

Dataset of 30 dynamic FFT vibration spectra collected from motor bearings operating under varied load conditions.

How this agent runs

Evaluates accuracy of bearing fault mode detection from spectral peak frequencies.

Challenge input

JSON containing peak frequencies (Hz) and amplitudes (g)

Claude Agents SDK

Advanced extended thinking capabilities for technical root-cause reasoning.

Evaluated output

JSON detailing root cause fault mode and estimated remaining useful life (RUL)

Checks for
  • Ensure fault_mode matches standard bearing defect classifications
  • Ratio of correct fault mode diagnoses across evaluation suite • target: 0.93 • range: 0-1
Proof of success
  • Benchmark: Predictive Maintenance Vibration Suite (PdM-VIB-2025)
  • Diagnostic Accuracy target: 0.93
  • 1 public reference case
Runtime evidence
  • Python execution harness
  • Python sandbox (unavailable on Versalist)
View technical recipe

Configured tools

Action Space
  • Claude Code · Required
  • Anthropic · Optional
Observation
  • Claude Code · Required
Policy Serving
  • Claude 4.1 Opus · Optional

Evaluation contract

  • valid_fault_type · Weight 1
  • diagnostic_accuracy · Weight 1

Recipe state

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

Frequently Asked Questions about Industrial Equipment Vibration Diagnostics Agent using Claude Agents SDK