projects

Designed and built the backend for a multi-agent AI system that analyzes patient lab results, diet logs, and intake data to generate personalized gut health reports. Event-driven job pipeline with async orchestration, vector search over medical knowledge bases, and structured LLM chains across specialized agents — each handling diet analysis, biomarker interpretation, or treatment planning.

Full-stack Databricks observability application with a K8s-based connector to the Cortex IDP, standardizing Databricks metadata ingestion and driving best practice adoption at scale. Deployed at ASMLGlobal leader in semiconductor lithography systems, enabling the production of advanced microchips.↗ Visit website and SephoraMultinational retailer of personal care and beauty products with 2,700+ stores worldwide.↗ Visit website to save +$1M/yr.

Proof-of-concept for a portable lakehouse solution using DuckDB to modernize legacy data architectures. Integrated with Unity Catalog for governed, high-performance analytics.

Industrialized an NLP model for HS code package classification at PostNLThe national postal service of the Netherlands, delivering 1M+ parcels daily.↗ Visit website. Built the backend ML and data infrastructure from scratch — €1M+ savings, Innovation of the Year award.