last updated · august 2026
now
These are the things with most of my attention at the moment. This page is a snapshot, not a promise.
what software should get to do
I'm thinking about useful AI that does not get in the way: tools that help people find context, leave a trail of reasoning, and remain understandable after the novelty fades.
three projects, only so many evenings
sastram
an AI-powered community space, currently taking most of the time.
protor
small improvements to the local-first web analysis workflow.
dripl
polishing the parts of collaboration that people actually notice.
away from the editor
Reading more slowly, collecting music, walking without a podcast, and trying to leave enough room for an idea to arrive unannounced.
not now
Not chasing every new framework. Not making a project bigger just because I can. Not pretending the unfinished parts are not the interesting ones.