Hey, Dylan here.

I've been deep in Claude Code for a while now, and lately I've been paying more attention to the decisions I make while I build.

When am I exploring vs just accepting the first thing that compiles? When am I forcing a decision I'm not ready to make?

I noticed I often don't know what I want until I have something to react to. So I built Cookoff, which gives the same spec to multiple agents so we can compare what comes back. And Omakase, which has Claude build every option so I can pick from real things instead of hypotheticals.

I also noticed I kept committing to answers before I understood the tradeoffs and the ramifications. So I built Deliberation. It surfaces perspectives and lets me think through the reframed problem a bit longer before I decide.

We're all building with AI now, so I wanted to share what's been working for us.

Hope it's helpful,

-d.

Dylan and the 2389 team