At Datafrik.co, I designed Snowflake star-schema models for the Northwind Threads and Streaming datasets, structuring raw, staging, and mart layers to support five downstream reports. I also built dbt models that reduced duplicated SQL by 30%.
I implemented custom data quality tests that caught six issues before they reached dashboards, and published dbt documentation and lineage. I delivered two end-to-end projects using Git and GitHub pull requests with CI/CD.
At Tween Technologies, I audited warehouse inventory data and documented missing and malformed values that affected sales and margin calculations. I also built a pandas cleaning pipeline that recovered all 1,000 rows with zero data loss.
