Deep technical screen

Data model and inference interfaces for fleet-scale reliability operations.

This screen is for engineering and architecture reviews: event schema contracts, inference loop execution, and integration patterns for consuming evidence and risk API outputs in downstream systems.

Core entity model

Primary records and relations in the NADIR control plane.

Entity	Key fields	Purpose
SignalWindow	vehicle_id, ts_start, ts_end, quality	normalized ingest unit for inference
DriftEvent	yaw_residual, confidence_delta, severity	actionable reliability event
InterventionRun	workflow_id, technician_id, status	calibration execution lifecycle
EvidenceLink	event_id, intervention_id, signature	proof chain for claims/compliance
RiskSegment	segment_id, risk_score, modes	simulation and OEM API payload

from nadir import FleetClient

client = FleetClient(api_key="NADIR_API_KEY")

for window in client.telemetry.windows.stream(fleet_id="fleet_mw_01"):
    residual = client.inference.estimate_residual(window)
    severity = client.inference.classify_severity(residual)
    if severity in {"high", "critical"}:
        client.workflows.dispatch_intervention(window.vehicle_id, severity=severity)

import { FleetClient } from "@nadir/sdk";

const client = new FleetClient({ apiKey: process.env.NADIR_API_KEY! });

const events = await client.compliance.alerts.list({
  fleetId: "fleet_mw_01",
  minSeverity: "high",
  include: ["evidenceTimeline", "recommendedAction"]
});

for (const event of events.items) {
  await client.workflows.dispatchCalibration(event.vehicleId, { priority: event.severity });
}

GET /v1/risk/segments?region=us_midwest&condition=night_rain&min_score=0.7
Authorization: Bearer NADIR_API_KEY

{
  "window": "rolling_24h",
  "segments": [
    {
      "segment_id": "seg_9192",
      "risk_score": 0.84,
      "failure_modes": ["lane_visibility_drop", "camera_vibration_peak"],
      "evidence_refs": ["ev_22431", "ev_22457"]
    }
  ]
}