It's been a week, but I'm still learning

So here's a random selection of the stuff I've been messing with

I'm really bad at this blogging thing, clearly. Check that drafts folder though, there's some half written gold in there. So in order to bump a particularly badly written post of the top of the masthead, I'm going to write a post about what I learned this year(ish). What year is it again?

Evangelizing dbt testing

I've been continuing my mission to get colleagues to embrace testing in our dbt project. We've got 2000+ models now, and the entire test strategy is unwoefully unloved, but we're working on it. I'm done talking about why "tests are essential guard rails, especially as our team grows and technical debt accumulates" because it's boring, and doesn't cut through. Can we make testing cool again? I'm not sure, but I'm trying.

The belated TM1 farewell message that no one will ever read

I decided a couple of years ago that I'd prefer never to work with TM1 again. I built some cool stuff with it but I found the developer experience to be, well, pretty poor. I still think it's an amazing tool for a certain type of problem, but I'm not going to miss it. I ended up really frustrated with the lack of focus on tooling that would make a proper CI/CD approach feasible. I have fond memories of working on TM1py though and I'm happy there are people still pushing smart ideas.

Embracing agentic coding

Isn't everyone? I was a bit late to the party but I'm enjoying using cursor at work. Morten, a former colleague, wrote this piece that I found pretty insightful. My main takeaway was that the garbage in, garbage out rule will hold true forever. Is that such a surprise? Adding strict rules and using agents in a targeted way brings dividends, but I still find the need to lecture Claude on a daily basis.