How I work

AI is my leverage, not my judgment

I work with AI tools (Claude and Claude Code, mostly) the way a strong operator works with a strong team: I decide what to build, what good looks like, and what's true, and I use AI to compress the time between deciding and shipping. Everything that leaves my hands is something I can defend line by line.

Concretely, this is what that looks like:

I automate real work

The Dune Lakes analytics pipeline is one command: it ingests a raw OwnerRez booking export, cleans out the linked $0 block rows, validates channel fees against actual fee data, and produces a formatted Excel workbook with charts. Rebuilding the analysis after a new export takes seconds instead of an afternoon. I've built similar automation around my own job search: resume tailoring, keyword matching against job descriptions, and application tracking.

I ship faster than one person should

NotTomorrow is a full-stack SaaS product with an AI planning engine and subscription billing, built and monetized by one person. AI-assisted development is the only reason that was possible on that timeline. This site is another example: I designed the structure and wrote the standards, and built it with Claude Code as the pair.

I keep the numbers honest

Speed is worthless if the output is wrong. The habits come from assurance work at EY: every figure traces to a source, estimates are labeled as estimates, and when the data can't support a metric (like true occupancy in the Dune Lakes analysis), I omit it rather than fake it. AI accelerates my work; it doesn't get to lower the bar.

The toolkit

Claude and Claude Code for analysis, automation, and building
Python (pandas) for data pipelines
SQL and Power BI for reporting
Excel when Excel is the right answer, which is often