Dennis Van de Vorst

Designed and implemented monitoring and alerting functionality for Stedin's MLOps platform, improving the stability and reliability of model predictions across the entire organisation.

Designed and built a completely new data lakehouse on AWS with Databricks, Airflow, and Airbyte, replacing the existing Microsoft SQL Server data platform and SageMaker ML environment.

Expanded PostNL's Data Lake v2 platform on AWS and migrated existing pipelines from their v1 platform to the new v2 platform.

Designed and delivered a two-day interactive AI/MLOps workshop tailored to the Efteling's infrastructure needs.

Delivered four ML projects with measurable impact: €25M in cost savings through predictive maintenance of smart meters, €350K/year saved and a more stable grid through grid loss forecasting, plus safer field work and winter outage-risk detection.

Built an AI-powered Twitter sentiment analysis platform using ROBBERT (Dutch BERT) to help emerging cultural talents monitor social media engagement.

Developed a scalable ML solution on Azure to monitor B2B chat messages for Code of Conduct compliance.

Wrote automated Python tests for embedded systems and designed Python APIs for equipment interfacing.

Built real-time telemetry, a CAN bus data logger and CAN analysis tooling in Python and C for TU/e's electric Formula Student race car, and designed its complete wiring harness.