About

What I care about most in data work is trustworthiness. A pipeline that moves data isn't finished until it validates that data; a forecast isn't useful until it's been tested against data it hasn't seen. That conviction runs through everything I've built: ETL pipelines with null handling, outlier detection, and distribution-shift checks; time-series models validated with holdout testing, confidence intervals, and residual diagnostics; and a cloud ingestion platform where data can't reach consumers until a human approver signs off.

My path into data engineering started with machine learning research. My bachelor's thesis — an SVM-based sentiment classifier for cyberbullying detection — meant annotating and reviewing 40,000+ text samples by hand. That experience taught me something most engineers learn late: model performance is mostly a data quality problem. I've been building systems around that lesson ever since.

I work primarily in Python and SQL on Azure (Azure SQL, Cosmos DB, Azure Storage, Key Vault), with .NET for backend services and Power BI for reporting. I also build with AI-assisted engineering workflows — this site was developed spec-first with an AI coding agent working under human review gates. I hold the Microsoft Power BI Data Analyst and Meta Database Engineer professional certificates.

I'm looking for an AI/ML engineering or data science role on a team that treats data quality and governance as engineering problems worth solving well.