MICHAEL HUSS

Architected an enterprise clinical data fabric integrating EHR, claims, and real-time telemetry, processing 10B+ records/month and reducing latency to near real time. Implemented a Delta Lake medallion architecture for 3x faster analytics/ML queries and cut Snowflake compute costs by 40% while enforcing HIPAA-compliant governance (RBAC, masking, lineage, audit logs).

Led development of a real-time personalization engine ingesting and processing 5M+ user events/hour using Kafka, Spark Structured Streaming, and AWS Kinesis, driving an 18% conversion lift. Built scalable ELT pipelines with Airflow, dbt, and Snowflake, migrated legacy ETL to modular dbt for 2x faster deployments, and improved observability to reduce MTTD/MTTR by 50+%.

Engineered a metabolic health intelligence platform ingesting IoT health signals via Kafka and GCP Pub/Sub, processing billions of events/month. Built GCP pipelines (Dataflow, BigQuery, Cloud Storage) and feature engineering with PySpark to improve chronic disease prediction accuracy by 25%, while reducing data inconsistencies by 30% through validation/anomaly detection and cutting release cycle t

Built an imaging analytics pipeline with PySpark, AWS EMR, and S3 for radiology imaging metadata, reducing processing time by 45%. Developed backend services and APIs using Java (Spring Boot) and Python, and improved production reliability with logging, monitoring, and alerting.

Developed a healthcare data lake platform with AWS S3, Glue, and Redshift, integrating multi-source datasets (claims, EMR, lab results) and processing terabytes daily via batch and streaming pipelines. Optimized Redshift performance with distribution/sort keys and query tuning to reduce execution time by 35%, and implemented data quality and lineage frameworks to improve trust in analytics outputs

Bachelor of Computer Science from the University of Central Florida (2014–2018).