Chris Dabel

Data engineer and architect. I build the data layer AI runs on, and I build with AI while I do it.

Twelve years of pipelines, warehouses, and the infrastructure behind production analytics and ML. Most recently the founding data engineer at a fintech, where I owned the whole path solo: ingestion, a Snowflake medallion architecture, orchestration in Airflow on AWS, and the governance that let an LLM product see only what each tenant was entitled to. The same platform fed the model: a labeled corpus in the millions of rows, a harness for scoring candidate models against each other, and per-call cost analytics so nobody had to guess what the AI was spending.

AI is a multiplier in my work, not a substitute for it. Agentic development in Claude Code, custom skills and MCP servers, RAG on Snowflake Cortex. This site was built that way and the source is public. Knowing where AI belongs in a system, and where it does not, is most of the job now.

I do my best work at the edge of a problem nobody has mapped yet. Hyperfocus is the engine. Writing is how I keep it honest: inhabit the ambiguity, find the structure inside it, cut everything that is not load-bearing, ship something people can use.

Before data I was an analyst: a news desk, then a State Department internship in Kinshasa, working in French. Translating between worlds is the oldest skill I have, and it is still the one the business question and the technical answer both need.

Looking for the next one: founding or staff-level data work, high ambiguity, high autonomy, with leadership that wants architecture shaped with them rather than handed to them.