Zach Gollwitzer

Lovable/Cursor/Bolt have created a "Home Depot moment" for apps It hasn't killed software development; it has simply opened the doors for non-professionals to participate in something previously inaccessible due to skill and capital-based barriers to entry. "But that app is terrible, wait until they deal with the reality of maintaining an app with real customers and scaling infrastructure..." I think this argument misses the point entirely. Very few people walk into Home Depot thinking, "I'm going to build my dream home and then immediately start a home-building company" People go to Home Depot to buy the things required to solve their small and idiosyncratic problems—painting a room, fixing the deck, building a house for their dog. I think the same thing is happening with apps—there's a level of joy AND utility in building an app that solves a personal and potentially, ephemeral need. Home Depot didn't displace professional painters nor did it put them in competition with the DIY'er attempting to paint a room in their house to save some money, learn a new skill, for the enjoyment, or all of the above. I think the tension we see today occurs when the DIY painter finishes their first room and immediately starts bragging to the world, "There's no need for professional painters anymore; Just go to Home Depot and do it yourself!" The definition of "app" is changing. While previously, consumers equated "app" to "highly professional and reliable", we're now seeing a wide array of "apps" flooding the market; some great, some good, and some, quite frankly—dangerous to consumers. I really think the only outcome here is consumers increasing their skepticism while subscribing to new apps and placing a higher value on the "origin" of the app. Durable apps will continue to capture the majority of the market. Just like the industrialization of farming led consumers to care a lot more about the sustainability and origin of their foods, I think this mass-production of applications will wake consumers up to the quality of the apps they're using; not just whether it looks good and works well. It will probably also lead to more stringent review processes in app stores :/ Nevertheless, AI code generation has opened up a creative outlet for thousands of people that will most certainly push existing apps to innovate faster. And for that reason alone, I think it's a huge win for everyone.

It’s funny how through all the cycles of worktrees, subagents, skills, ralph loops, lobsters, and context engineering, a good ole default claude code instance + nvim/tmux + some hard human thinking still seems to accomplish more in the long run.

The bottleneck has so quickly moved from code generation to code review that it is actually a bit jarring. None of the current systems / norms are setup for this world yet.