Built to scale to a major-company-sized portfolio

Every telemetry source (OBD/CAN, GPS, driver app, claims, weather) lands on an append-only event bus (Kafka/Kinesis-class). Producers and consumers scale independently, so onboarding a new OEM/telematics vendor or a new insurer's claims feed never touches the scoring path.

SafeGuard, RapidRate, ClaimSense and PredictEdge (lib/engine in this technology) are pure functions over a feature vector. In production they run as horizontally-scaled stateless services behind an API gateway — scale is linear with fleet size, not with model complexity.

Raw signals are aggregated once into a shared online/offline feature store. The dashboard, the pricing engine and the claims-fraud model all read the same driver/vehicle/route features, so scores stay consistent across every surface.

Every record carries a tenant (insurer/broker/fleet) and policy-scope key from ingestion through to the API. Row-level isolation plus per-tenant encryption keys let IR Fleet serve a major company's captive, a broker panel and a direct fleet client from one platform without data crossing boundaries.

Every score already carries its contributing factors (see the behavioural breakdown in the live dashboard). That is not a UI add-on — it's how the engine is written: weighted, named sub-scores, never an opaque black-box output. This is what actuarial and compliance sign-off requires.

The engine recommends (coach driver, adjust premium, trigger maintenance, investigate fraud); it does not auto-execute against policy or bank systems. Every action routes through the insurer's existing underwriting/claims workflow and audit trail.

Odometer reading and condition photos are captured before a vehicle goes to a driver, timestamped and stored against that vehicle's record. This is the same evidentiary logic as the ClaimSense liability-dispute reduction — an objective initial-state record beats a disputed memory when a damage claim is filed.

Each vehicle's forward claim rate becomes claim probabilities over 30, 90 and 365 days, a per-category loss forecast and wear-driven deadlines for brakes and battery. Every recommended action is valued by re-running the same risk model with that action applied — the drop in expected loss is its benefit — so a fleet manager sees which vehicles to act on first and what each action is worth. Category mix, severities and action costs are documented assumptions that the pilot calibrates against the insurer's own loss data.

ClaimSense and RapidRate connect through a single operator-only console with an API URL and key — encrypted at rest, never shown to customers. Responses are validated against a published contract; if an engine is slow, down or returns something unexpected, the built-in model answers for that vehicle so the platform keeps working and the failure is visible only to the operator.

A fleet manager can upload their whole fleet as CSV, Excel or JSON instead of adding vehicles individually — column or key names are matched to our fields by word-overlap similarity, not exact string matching, with a manual review step before import since every provider's export looks different. For continuous updates, the same field-matcher runs unattended behind a plain JSON endpoint (POST /api/admin/units/ingest) that any polling script or telematics relay can push to; records are upserted by plate, so a vehicle reported every few minutes updates in place instead of piling up duplicates. Either path runs the identical validation and scoring as a one-by-one entry.

Claims-history + one live telemetry source (GPS or OBD) on a bounded pilot fleet. Validate score calibration against a major company's actual loss data.

Add remaining data layers, stand up the feature store and event bus for full fleet volume, integrate RapidRate into underwriting workflow.

ClaimSense fraud detection live on claims intake, PredictEdge portfolio reporting for reserving, broker-facing risk-improvement reporting.