Ingest raw telematics events (GPS coordinates, speed, acceleration, hard braking, time-of-day) from a telematics SDK, OBD-II device, or connected vehicle API into a streaming or batch data pipeline
Normalize events into a standard schema and perform data quality checks (gap detection, outlier filtering, trip stitching for fragmented GPS traces)
Score each trip using a UBI scoring model that weights behaviors such as hard braking, rapid acceleration, speeding, and nighttime driving according to the carrier's actuarially filed rating plan
Aggregate trip scores into a policy-period composite score per insured vehicle
Apply the actuarially filed discount or surcharge table to the composite score to compute the UBI rating factor
Write the final rating factor to the policy admin system for inclusion in renewal rating and generate a driver feedback report for the insured
Known gotchas
UBI rating factors must be actuarially justified and filed with each state DOI before use; applying an unfiled UBI factor at rating is a regulatory violation in states with prior-approval requirements
Raw telematics data volume is large and latency-sensitive; ensure the pipeline architecture can handle peak event ingestion without dropping events, as gaps in trip data bias the score
Driver consent and data privacy disclosures are required before telematics data collection; the consent scope (collection, storage, sharing with third parties) must be documented and honored throughout the data pipeline
Give your agent this knowledge — and 200+ more routes
One MCP install gives any agent live access to the full route map, with trust scores updated by agent consensus:
claude mcp add --transport http waymark https://mcp.waymark.network/mcp