I love ideas, truth, and beauty.
I'm 35.7 years old. I spend my days helping teams navigate complex engineering projects while researching information design: the art and science of making information understandable. This website is my notebook, for ideas, articles, projects, and experiments.
I work in Project Controls on some of the energy sector's largest EPC projects: pipelines, plants, cross-border infrastructure. My job is officially schedules, cost, and reporting. My real job, the one I chose for myself, is turning overwhelming complexity into something a project team can look at and know exactly what to do next.
EPC: Engineering, Procurement, Construction. Schedules on projects like these can run to thousands of linked activities.
A good diagram doesn't just show data. It moves an entire project team in the same direction.
Most of that complexity gets buried in procedures, thick documents, and slides full of text nobody reads. I got obsessed with the alternative: information design that actually moves a team toward a decision. Over the years I built an internal practice around it, rules, templates, visual standards, mostly locked inside the companies I built them for.
Then I noticed the same failure somewhere else entirely. AI agents and LLMs, for all their capability, are remarkably bad at this exact problem: generic, bold, one chart type for every question, no judgment behind the choice. There was no protocol telling them any better.
So I'm doing the same thing twice. I'm publishing papers that explain the principles behind good information design under complexity. And I'm building the Open Visualization Protocol (OVP): a deterministic, open library where every chart type and every visual language is spec'd precisely enough that a human or a machine produces the exact same correct result. Years of practitioner notes, given away.
OVP today: 35 chart types built, 16 design languages, one frozen spec.