Data Lifecycle & Documentation

Detailed explanation of simulation, experiment, and inference data structures in the ALP system.

Sim Request

Simulation

→

Exp Run

Core

→

Sensor Data

Hardware

→

AI Inference

Intelligence

1. Simulation Data

Simulation requests represent the triggers for autonomous code execution or virtual experimentation.

Field Type Description
snippet_id ULID ID of the code snippet to be executed.
input_params JSON Parameters passed to the simulation engine.
execution_result JSON Output data from the simulation run.
status String Current state: PENDING, RUNNING, COMPLETED, FAILED.

Example Input (JSON):

{
  "target_property": "conductivity",
  "iterations": 10,
  "constraints": { "temp_max": 200 }
}

2. Experiment Run & Sensor Data

Actual physical or high-fidelity virtual experiments recorded in the laboratory.

Field Type Description
parameters JSON The specific recipes/conditions used in this iteration.
reading_type String The type of measurement (e.g., UVVIS, CONDUCTIVITY).
reading_value Double The primary numerical result from the sensor.
file_path String Path to large data files (CSV, Spectra) if applicable.

3. AI Model Inference

The intelligence layer where models predict performance or optimize the next experiment steps.

Field Type Description
inference_type String The goal (e.g., PROPERTY_PREDICTION, NEXT_STEP_OPTIMization).
confidence_score Double Reliability score of the prediction (0.0 to 1.0).
input_features JSON Combined experimental data used as model input.
predicted_output JSON The AI-generated prediction or suggestion.

Example Inference (JSON):

{
  "predicted_value": 4.52,
  "unit": "S/cm",
  "suggested_next_recipe": { "ratio": 0.8 }
}