Fleet Complete: an ML platform for 600,000 vehicles
Lead Product Manager, Data & ML Platform · Jan 2020 – May 2021 · Toronto
Context
Fleet Complete is one of the world's largest telematics providers. I was brought in to lead the data platform's transformation, from a company that collected IoT data to one that monetized intelligence from it. I authored the company's data strategy and built the machine-learning platform to execute it.
What I built
Centralized IoT data from 600k+ vehicles into a 0→1 AWS-powered platform (S3, Glue, EMR, Athena, Kafka, SageMaker), converting static reports into dynamic, AI-driven insights.
Cut model training cycles by 99% (weeks → hours) with reusable data-scientist onboarding templates and CodeBuild pipelines orchestrating preprocessing, labeling, training, evaluation, and deployment.
Launched an external ML model-inference pipeline so partners could deploy models on Fleet Complete data, powering real-time AI predictive maintenance (publicly announced Pitstop partnership) on enterprise accounts representing a $300M ARR opportunity.
Oversaw petabyte-scale data migrations, designed the anonymization framework for secure cross-company data sharing, and modernized architecture, cutting latency, infrastructure costs, and data-related support tickets 4x.
Work contributed to a government-backed AI supply-chain consortium (Scale AI / OCE) project on predictive fleet maintenance.
"Under his leadership, the data platform team became an ownership-driven team able to deliver a cloud-based platform enabling adaptability and efficiency for the organization."