About

Notes on what happens when I try to do new stuff.

I spend most of my time trying to understand why systems behave the way they do once numbers hits real life datasets. That habit comes from years of working with data, automation, and code where inputs are super messy, hypotheses semi-testable, and small Ns turn into large problems.

When ChatGPT became the cool thing, flooding everyone’s feeds and inboxes, I wanted to learn more. And writing is one of the best ways of learning. So I started Let’s Just Do AI as a place to think out loud and log what happens I try things out myself.

Most of what appears here begins with curiosity and a good dose of skepticism. A new model comes out, a new super-cool-amazing agent shows up, or a workflow-tool-whatever promises to save years of work. I run it. I test it the best I can without spending a fortune in API calls. Sometimes it works, sometimes it doesn’t. The interesting part usually lives in between, and that is what I write about.

I am a computational scientist by training, which mostly means I am used to stare at a screen for a very long time trying to make sense out of large datasets. I live somewhere in between data analytics, oncology, data science, and, lately, applied AI. Let’s Just Do AI reflects that unpolished version of my thoughts that I type out when I have the time to play with new cool stuff I come across.