Data is only as good as what you do with it
- I transform data with dbt, SQL, Python, Scala and Spark
- I live in the clouds: AWS, GCP, Databricks, Snowflake
- I orchestrate workflows with Airflow and Prefect
- I load data with dlt, Fivetran, and Stitch, from raw sources into the warehouse, reliably
- I build reliable lakehouses on Delta Lake and Apache Iceberg
- I design data models that scale: Kimball, Data Vault, Medallion, SCD Types
- I build source → staging → intermediate → mart layers with clear contracts between them
- I define metrics once and trust them everywhere using dbt Semantic Layer and MetricFlow
- I write tests because data quality matters: schema tests, custom tests, Elementary, Anomalo
- I track column-level data lineage so nobody asks "where does this number come from?"
- I treat analytics code like software: CI/CD, versioned transformations, structured code review
- I write PR descriptions that explain the why, the impact, and what to watch out for, not just the what
- I document models, sources, and exposures so teams don't need me to answer their questions
- I enforce data contracts between producers and consumers to catch breaking changes early
- I catalog and govern data with DataHub, Collibra, and Unity Catalog so data is discoverable and trusted
- I optimize until things run fast and cost less: query profiling, clustering, incremental models
- I experiment with DuckDB & ClickHouse for fast analytics
- I make dashboards that people actually use: Looker, Tableau, Superset
- I monitor everything with Grafana, Elementary, Anomalo
- I translate business needs into data solutions
- I do graphic design on the side (Adobe suite enthusiast)