Engineer, analyst, occasional economist.

I started in economics — not the kind that wins Nobel prizes, but the applied kind: regression tables, policy analysis, late nights with Stata. What I found, slowly, was that the most interesting questions weren't in the models themselves but in the data underneath. Messy, incomplete, contested data. I wanted to get closer to it.

That pull led me into analytics engineering. I learned SQL deeply, then Python, then what it meant to build a pipeline that survives contact with production. I worked on data infrastructure for e-commerce platforms — Kafka streams, Airflow DAGs, warehouses that actually held up under load. It was unglamorous and I loved it.

The next chapter was machine learning: not the research kind, but the it has to work on Monday morningkind. Lead scoring, demand forecasting, churn prediction. I built feature stores, owned model deployments on SageMaker, and learned the hard way that a model that doesn't ship is just an expensive science fair project.

Now I work at the intersection of AI and production systems. LLMs, multi-agent pipelines, RAG architectures — the field moves fast, but the fundamentals haven't changed: understand the problem, ship something real, measure what matters. I'm still chasing those hard questions, just with better tools.