Every couple months I do a race where I have some agents go off and build a feature or fix a bug while I do it myself in Cursor. The time I spend reviewing and fixing the agent's work always end up being longer and more painful, which is I haven't switched over to an "agent command-center" style of software dev. I do kick off worktree agents here and there throughout the day to make minor changes that come up while I'm working on a larger branch. But those are side quests while I work on the main thing.

Cursor's Composer 2 model is performing much worse for me than Composer 1 :( I feel like Composer 1 really hit a sweet spot for me between speed and quality. For me the bottlenecks for coding with AI are: - understanding all the code that the model wrote - testing changes Composer 1 really helped with the first because it could blast out small amounts of code that I could quickly review without my brain getting bored and context switching to something else. I feel like I'm an outlier in that I'm trying to stay heavily involved in the dev flow rather than having a multiple agents work on long tasks and then coming back in cold to review their work. Is anyone else using smaller quicker models in this way?

Do we have a NIP/tag that says to a relay "only serve this event to the authenticated author or p-tagged recipient?" This behavior is mentioned in NIP-17 and NIP-9a and probably makes sense in a lot of cases, and I want it for my Shamir's Secret Sharing NIP.

Just had an hour long video call in flotilla (the video part is still in dev, not released yet). The call quality was actually really impressive, better than Jitsi or Keet I would say. Props to Livekit for the killer open source WebRTC toolkit.

I'm going to be hanging out in in the "Voice Chat" room in Flotilla today in case anyone wants to jump in and try it out! Here's an invite to the space: https://app.flotilla.social/spaces/meta.spaces.coracle.social/trarghstroyno6 View quoted note →

TIL the fairphone people make airpods with replaceable batteries:

Fairphone

Fairbuds, the repairable earbuds | Fairphone

Meet the world's most repairable premium earbuds. Replaceable batteries inside buds and charging case.

This blog post about the end of the American Empire has been living rent-free in my head since Saturday. It's quite long but it touches a lot of ideas that have been rolling around in my head like: how quickly will the American Empire fall apart, is it worth trying to reform the current system, how can we molt into better forms of governance through it? The idea that territorial sovereignty as a concept is on its way out is totally new to me but very intriguing.

A Farewell to Empire

Post-Westphalian Imagination in the 250th Year of the United States

If you're curious what I've been working on for the past month, I published my quarterly report for Opensats on my blog: https://mattlorentz.com/2026/03/30/opensats-q1-grant-update.html

Lifetime iOS user 3 days into using GrapheneOS as my daily driver. AMA.

This guide was so useful. Especially the app recommendations as a user coming from iOS. View quoted note →

I'm up for this too! View quoted note →

Hello funny square phone.

I'm planning out my trip to DWeb camp in Berlin this year. I've never been to Germany. Does anyone have recommendations of things to see while I'm there or places to stay? Any advice on traveling in Germany?

One part of Star Wars that always seemed unfuturistic was how disconnected all the tech was. Like they talk to each other on radios and carry data around on chips and pay with cash a lot of the time. But watching Andor today it doesn’t seem so silly. Like I could totally see a world where AI accelerates offensive capabilities so quickly that as a defense we just turn off the internet and computers become simpler tools again.