**course material for adults who swear**

A little known fact is that RStudio rendering is powered by users’ electromagnetic fields (i.e., “good vibes”) and the exodus to Positron has severely limited its ability to compile code. #rstats

Modules + Claude code for simple but labor intensive edits across files

Tbh I relate to that big yellow spike of with mad uncertainty around age 30. 🤣

I think if you’re curious and truly care about problem solving you might have a ~temporary~ feeling of closure or a premature commit. But you will keep iterating (opening/closing) as you explore the problem space and how it works, validate the throughput, and improve the methods. Stay curious!

I would use cosine in stringdist. If you have lists of job descriptions (from two sources with each idx being a similar job), you can use my package samesies. dylanpieper.github.io/samesies/

So cool! Any intros or docs planned for helping people familiar with the futureverse make the leap to marai?

One cool thing you can/should do is sample from priors only, and plot the distribution of the actual quantity of interest (ex: risk ratio). I find this very useful. This is actually super easy with brms. arelbundock.com/posts/margin...

The worst is when you write in active voice and then someone tries to edit all of it back into passive. Old habits die hard and the good fight continues.

You could use surveydown and provide the LLM with the package docs