Week 8, 2026 — hello Claude
I went to see the exhibition L’empire du sommeil at Musée Marmottan Monet, which had some nice pieces but felt like an unserious attempt at tackling a rich theme like sleep. It didn’t explore dreams, day dreams or napping at all. And there were a few too many pieces that were “man/god looking at nude, sleeping female“ to which I had a heightened awareness of, as our visit was the same day that former prince Andrew was arrested. Still, a museum visit and light dinner afterwards is such a wonderful way to spend a weekday evening and I’m determined to repeat it soon.
A big week for Emotions at Work, as we kickoff a multi-month engagement with a new client. I was reminded yet again how energizing it is to spend time with open-hearted clients that seek partnership to make a dent in their aspirations.
•
I finally took a proper plunge into vibe coding, and got from point A - hey here’s an idea to point B - hey here’s the URL of my app, try it out. Yes, I got as far as hooking up a Github repo, Vercel, and Supabase so that I can deploy web apps with a database backend. It’s been a lot of fun and I’m liking the learning curve. I think I can become much better, very quickly, and then see how useful that is…!
I worked on two apps:
An learning app that helps increase fluency with a Gestalt framework that I’ve been having retention issues with—I generated exercises, and designed a chatbot that acts as a teacher. It’s laser focused on helping you learn the material, and doesn’t try to analyze you or “support” you, which is an issue with a generic LLM experience. My goal is to make a couple of single-purpose learning apps under the umbrella of Gestalter for me and my peers. This is a very good start!
A whiteboard tool for a workshop activity that we often run at Emotions at Work, for which I have not been able to find a sufficient workaround for in Miro or Figma. There are still technical challenges to be overcome but I’m more and more optimistic that I can build something we can use ourselves. Once I’m there, we’ll explore the possibility of releasing it as a paid service. This is a much more ambitious app than a Claude API wrapper, and I’m mindblown at how far I got just sitting on my couch on the weekend.