About

Welcome to “Sometimes Models Just Do Things,” a blog where I document my temporary escape from PowerPoint decks and Partner meetings to wade into the strange waters of artificial intelligence.

As a 41-year-old consulting partner with five children under nine, I’ve traded quarterly earnings reports for prompt engineering, mostly conducted in five-minute intervals between toddler negotiations.

This blog chronicles my journey after what I can only describe as a “Deep Research moment” – that instant when I couldn’t un-see that reinforcement learning, now unleashed on knowledge work with LLMs as the transmission mechanism, will produce models smarter than me within a year or two.

I’ve always been a tech nerd in my spare time – building LEGO robots with the kids, coding up unnecessarily complex home automation systems, and explaining to my eight-year-old why his AI-generated homework isn’t fooling anyone.

But now I’m bringing that enthusiasm into the firm, partly out of fascination and partly out of the professional self-preservation that consulting Partners develop after years of selling digital transformation to clients who still use Excel 2007.

All analyses are conducted one-handed on a mobile phone, typically while hiding in the bathroom or pretending to pay attention on conference calls. Efficiency isn’t just a buzzword here—it’s the only way this blog gets written at all.

Join me as I document what happens when someone with too many children, too little time, and just enough corporate authority decides to see if AI can save us all from ourselves. Results not guaranteed.