Data platform engineer at AB180. Feature platform and inference serving for an ad product. dbt/Snowflake on the training side, Python and Go services on the serving side.
Where I've been
- AB180 Β· 2023.04 β now β data platform (dbt, Snowflake, Airflow) and the ML feature platform. Reporting backend before that, and a stint doing data business development on the side.
- ETRI Β· 2021.03 β 2023.02 β integrated authn/authz for an open data hub. Keycloak, Kafka, Debezium CDC. Two first-author KCI papers and an M.S. from UST came out of it.
What I do well
- Finding defects nobody reported. Chronic undercounts in customer-facing numbers, a warehouse burning credits with no query behind it, a training pipeline that referenced its own downstream report.
- Turning the fix into a check, so the same thing can't come back. Both projects below started that way.
- Counting the cost before building. I've cancelled my own designs on that basis.
Making
- dbt-plan pypi forks last commit β
terraform plan, for dbt. Warns about destructive DDL and broken downstream refs beforedbt run. Static analysis of compiled SQL, no warehouse connection.pip install dbt-planΒ· listed in awesome-dbt - zero-shelter β turns scanner output into a short list of what to fix now. TypeScript, no LLM at runtime.
- autopilot-skills β mission runner for AI coding agents. Risk-tiered diff limits and a pre-execute deny list.
- lightweight-kubernetes-sandbox-cli β k3s cluster setup and ops CLI, from the ETRI years.