The article is a retrospective on some of the things some of the things that helped and some mistakes that slowed me down. The biggest lessons for me were:
1. Loss of focus.
2. Waiting too long before releasing.
3. Distribution turned out to be much harder than building.
4. AI significantly accelerated development but didn't solve the distribution problem.
5. Some decisions that looked like over-engineering ended up paying off.
Hope it's useful to anyone building something on the side.