Gary Grinko

I've built and optimized enterprise data platforms for Brosser clients across healthcare, logistics, fintech, and retail. My recent work spans Microsoft Fabric, Databricks, Delta Lake, Snowflake, dbt, and real-time streaming.

For a national retailer, I rebuilt an ad-hoc Databricks environment into a governed Delta Lake Medallion architecture, reducing monthly compute spend from roughly $150K to $97K and Gold-table query times from 15–20 minutes to under one minute. For a logistics provider, I architected Kafka-to-Snowflake streaming ingestion handling about 50K messages per second and replacing 24-hour batch visibility with sub-minute access.

At Comcast, I migrated roughly 2 PB from on-premise Hadoop to AWS and rewrote legacy workloads in PySpark on EMR, cutting batch runtime from eight hours to three. I also built telemetry pipelines processing 15–20 TB daily and mentored engineers on PySpark practices and code review.

Earlier at Bristol Myers Squibb, I improved clinical-trial ingestion, data warehousing, and pharmacovigilance reporting for more than 100 business users. I bring 13 years of hands-on data engineering across Azure, AWS, lakehouse platforms, streaming, data modeling, and BI.