I write these while I’m learning the topic myself — something from work, something I picked up for fun, or notes from a talk, conference, or course I want to revisit later. Writing it out in detail, the way I actually worked through it, is how I make sure it sticks. If it ends up useful to you too, that’s a bonus. I’m still learning, so cross-check anything before you rely on it — I can get things wrong.
Whether the problem involves physical systems, genomic data, or text, my approach stays the same: define the question precisely, work out its structure, build a reproducible computational model, and check it carefully against data. This page collects that work — statistical genetics, machine learning, natural language processing, high-performance computing, and scientific programming — from the math behind a method to the pipeline, the debugging, and the code that actually runs.
