Notes on R, statistical software and whatever else I find worth writing down.
Four ways to put an LLM next to R, how to keep trial data away from it, an analysis that looks right but is wrong, and how to use these tools without outsourcing the learning.
How I built a local RAG system around the MAMS R package in R: embeddings, retrieval and a Shiny chat interface, all running on a laptop.
browser(), debugonce(), traceback() and recover(), with examples in plain scripts and in Shiny apps.
Why 0.1 + 0.2 == 0.3 is FALSE in R, how much memory integers and doubles take, and where partial matching of names can catch you out.
Background jobs in RStudio, profiling with profvis, debugging with browser() and parallel loops with foreach.