So, making a real time collaborative tool from scratch there was a lot of deep work in things like sockets, queues, memory structures, etc. That's just what needed attention.

Right now I'm building a memory app for AI.

So far I haven't gone deep on the swift side of things. But I have spent a lot of time up front working on domain driven design, screaming architecture, and module based design.

I think with AI the polish is in deep QA, design taste, running trace logs and constantly asking it to do html diagrams of the architecture and such.

ngl I miss "knowing what's going on" at the code level, but it feels like not a good use of my time to go into those weeds.

When I look at the code it seems well structured due to screaming architecture, and module based design.