Connected-Fleet Telematics Platform for an Automotive Services Company

The Challenge

The company's maintenance business was reactive by design: vehicles came in when something broke or when a fixed service interval said so. Fleet customers were asking for more — visibility into where vehicles were, how they were being driven, and which ones were about to fail — and competitors with telematics offerings were starting to win those conversations.

The company had OBD-II telematics devices picked out and strong fleet-operations know-how, but no software organization capable of building a real-time IoT platform. Devices would stream position, engine, and diagnostic data around the clock from tens of thousands of vehicles; the platform had to ingest it reliably, make sense of it, and surface it in a product fleet managers would actually use. They wanted one accountable partner to take it from architecture to launch.

What We Did

Dillo delivered the platform as an outsourcing engagement — an architect, four engineers, and a QA engineer — with fleet-manager interviews shaping the product from the first sprint.

• Built device ingestion on AWS IoT Core, handling connectivity, authentication, and over-the-air configuration for the OBD-II fleet devices, with store-and-forward for cellular dead zones.
• Streamed telemetry through Kafka into TimescaleDB, storing high-resolution time-series data — location, speed, fuel, engine diagnostics — with continuous aggregates for fast fleet-level queries.
• Implemented driver-behavior scoring in Python: harsh braking, rapid acceleration, excessive idling, and speeding events rolled into per-driver safety and efficiency scores that fleet managers use in coaching programs.
• Developed predictive-maintenance models on diagnostic trouble codes, battery voltage trends, and usage patterns, generating alerts that route vehicles to service before failures strand them.
• Delivered a React fleet portal with live maps, vehicle health, driver scorecards, geofencing, and maintenance scheduling integrated with the client's service-center network.
• After GA, transitioned to an augmented follow-on team of three engineers who now own the roadmap — new device types, customer integrations, and model tuning — inside the client's product organization.

The Results

• 25,000 vehicles connected across the client's fleet customers, with onboarding of new fleets down to days.
• Customers using driver coaching and idle-reduction insights report around 18% fleet fuel savings.
• Predictive-maintenance alerts cut unplanned vehicle downtime by about 30% for participating fleets.
• The platform reached general availability in seven months, and telematics is now part of every major fleet-contract renewal conversation.
• The embedded follow-on team keeps shipping — the client gained a product line without building a software division from scratch.

Figures reflect outcomes reported for this engagement; they are project results, not audited benchmarks.

Tech Stack

Services Used

Software Outsourcing (end-to-end platform build) followed by IT Staff Augmentation for the embedded follow-on team.