Welcome!
This is David Wetterau's blog. I am a Software Engineer in New York City. I previously worked at Cursor on product, infrastructure, and training. Previously I worked at Airtable on AI and core product. Many years before that, I worked at Dropbox on some distributed systems.
Posts
-
A beautiful fall in New York City
I had the privilege of taking a break between my job at Cursor and my next role during the fall. Historically I have always gone straight from job to job, mostly to keep health insurance coverage (thanks America), and because I generally fear the unknown of not knowing what to do with my time without going to work every day. But this break taught me that even without work the world still goes on, and I can easily find things to do.
-
Tasky: State for my agents
While working at Cursor, I frequently juggled (and lost track of) many cloud agents for my coding tasks throughout the day. Often each agent would culminate in a PR, but not always. Sometimes I would have several agents drive a PR to completion, and sometimes I would produce multiple PRs from the same agent. It was common at the time, less so now, to experiment with products that would organize agents on a Kanban board, so that you could see more easily which agents were running, needed attention, and so on. But because the underlying task to agent relationship was not 1:1, I found these visualizations unhelpful.
-
AI Coding Summer 2026
The way I wrote code completely changed around the beginning of this year, which I wrote about some back in March. I noticed it first on side projects - where I found that fully vibe-coded apps were just as good as my old side-project attempts, as long as I code reviewed the parts such as the interfaces and data models. Anything outside of those it stopped making sense to sweat over, since a full re-implementation is so easily attainable.
-
Coding agents are no longer interns
Last year, I wrote about how LLM coding models at the time were capable of autonomously working at small, simple tasks, like those that you would scope out and assign to an intern. Since I wrote that about 9 months ago, both the models and my usage of them have evolved dramatically.
-
2026 Goals
My historical success rate on New Year goals is not very high, but over the years I’ve made some pretty great improvements to my life from them. Maybe by writing them down publicly, I’ll have a better chance of following through this time! If not, it will at least force me to reflect on them if they don’t work out.
- •
- 1
- 2