Pydantic AI Predictive Motor Maintenance Agent with Vast.ai Model Runtime
Vibration and motor temperature telemetry often signal impending Motor Control Center (MCC) failures before physical trips occur. Build a type-safe Python agent using Pydantic AI and run inference via Vast.ai GPU runtimes to process high-frequency motor vibration spectra. Success is achieved by predicting motor failure windows within a 24-hour error margin while outputting strictly validated Pydantic models.
What you are building
The core problem, expected build, and operating context for this challenge.
Construct a type-safe Pydantic AI agent executing on GPU infrastructure via Vast.ai to process high-frequency FFT vibration data and predict motor failure windows.
How work is evaluated
Evaluates remaining useful life (RUL) estimation accuracy and structural validity of motor telemetry analysis.
Challenge datasets
Public datasets, plus the private datasets you have access to.
How submissions are scored
These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.
pydantic_schema_valid
Ensures agent output strictly passes Pydantic BaseModel validation.
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
rul_mae_hours
Mean Absolute Error in RUL prediction hours compared to ground truth. • target: 12 • range: 0-100
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
What you should walk away with
Define Pydantic schema models for high-frequency motor vibration FFT spectra.
Build a Pydantic AI agent with explicit dependencies and typed output guarantees.
Configure cloud GPU deployment on Vast.ai for fast spectral feature inference.
Calculate Remaining Useful Life (RUL) estimates based on bearing fault frequencies.
Reference links and supporting material
Vibration sensor telemetry records including peak acceleration, RMS velocity, and bearing geometry parameters from motor control centers.
Challenge setup
Evaluates remaining useful life (RUL) estimation accuracy and structural validity of motor telemetry analysis.
Challenge input
JSON object with motor_id, peak_acceleration_g, and peak_frequency_hz
Pydantic AI
Required framework for type-safe Python agent development.
Vast.ai
Cloud GPU infrastructure for high-throughput spectral analysis models.
Evaluated output
JSON object with health_status, estimated_rul_hours, and bearing_fault_detected
- Ensures agent output strictly passes Pydantic BaseModel validation.
- Mean Absolute Error in RUL prediction hours compared to ground truth. • target: 12 • range: 0-100
- Benchmark: PdM-VibrationBench-2026
- Rul Mae Hours target: 12
- 1 public reference case
- Python execution harness
- Python sandbox (unavailable on Versalist)
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
- Pydantic AI · Required
- Vast.ai · Optional
- Pydantic AI · Required
Evaluation contract
- pydantic_schema_valid · Weight 1
- rul_mae_hours · Weight 1
Metrics
- rul_mae_hours · Target 12
Benchmarks
- PdM-VibrationBench-2026
- energy
Reference cases
- 1 public reference case
Workflow tasks
- Task 1
Input: JSON object with motor_id, peak_acceleration_g, and peak_frequency_hz
Output: JSON object with health_status, estimated_rul_hours, and bearing_fault_detected
Execution harness: Python
Evaluation tools
- Pydantic AI · Agent Framework
- Vast.ai · Inference & Model Runtime
Recipe freeze marker
No recipe freeze time is recorded.