One process that can work:
Step 1. Brain dump into a doc (consider using dictation to get more thoughts down faster)
Step 2. Have an LLM give it structure & progression. You are ordering your thoughts for readability, so you'll probably want to throw it away. You're still refining your thoughts at this stage.
Step 3. Take the LLM output as a starting point, or write an outline from scratch. Flesh it out into a first draft
Step 4. simplify: cut words, swap big words for small words, etc.
Step 5. Repeat step 4.
LLMs bridge the gap from word-vomit to structure. You should be willing to throw away what you get from the LLM.
At least 30% can always be cut. It's amazing how much can be trimmed without losing the intent.
I feel like the process of editing my own stuff is at least as important as getting it down. That's when I go back through it and realize all the conclusions I leapt to, things I didn't thoroughly consider, and other flaws. I think people really undervalue writing as a focus tool. But maybe that's just me, YMMV.
I think the same thing about a lot of code, too. Sometimes you really are just hammering out boilerplate. But a lot of times even writing test code is a great opportunity to realize the main code could be improved. But the LLM probably won't tell you that.
Expand, condense, compress, expand, compress.
Then somebody comes along and asks about something specific, so you expand.
Finally they use an LLM to compress after their pet idea was satisfied.
It’s crazy really, we’re just on a never ending rollercoaster here.