One thing I really agree with dhh on (and something I’m genuinely looking forward to!) is the resurgence of native apps.
Cross-platform became so dominant because the same small dev teams had to support users across multiple platforms. Practically they either had to build ONLY for the web or use some kind of web-tech-based framework like Electron.
Agentic coding changes that constraint at a very low level. Teams are no longer limited by whether they have someone who knows how to build for iOS, Android, macOS, etc., or whether they have the bandwidth to maintain multiple native codebases. The cost of supporting multiple platforms starts to fall dramatically.
And I think this is closely related to the thesis behind Omarchy also. Setting up and customizing Linux is historically a lot of friction. If you wanted a system that was personal i.e. NOT (Mac OR PC) you often had to be comfortable troubleshooting obscure issues and spending hours getting everything just right. Don't get me wrong: a lot of people enjoy that friction. But it sucks at so many levels.
Agents lower that barrier. You can want a highly customized system without necessarily becoming an expert in every layer of the stack. If there’s some tweak or workflow that isn’t exposed through a settings menu, you can increasingly just describe what you want and have the system help you build it.
So I think the interesting shift is that software becomes less constrained by what the average team or individual programmer has the bandwidth to build and maintain. That could make both native apps and highly personalized computing environments much more viable again instead of the web-slop that we've become accustomed to over the past 20 years. I really see parallels between how the Mac empowered creatives in the 80s and agentic coding today.
As much as i'd love for this to be true, it is not. Teams need to support their native apps, just because you prompted your way into some code doesn't mean your team is able to support it. The constraints are the same, if no one in your team knows how to build an maintain a native iOS app and their core knowledge is TypeScript and web, vibing it with agents won't get you past your first release (and not even that)
I don't really mean “give a team of TypeScript devs an agent and suddenly they can ship and maintain a production iOS app.” Expertise and ownership still matter, especially for something you have to support for years.
I'm talking more about agents as a force multiplier for teams who do have engineering experience. They reduce the amount of specialized knowledge, time, and coordination required to get something built and maintained. The constraint gets meaningfully weaker.
An experienced engineer who understands software architecture, debugging, product constraints, etc. but isn't an expert in every native platform can increasingly use an agent to bridge some of those gaps. That's very different from vibecoding your way into production.
fwiw I think we're still very early in the process of re-thinking about how teams are organized and what skill sets they need.
Really interesting experiments that are actually geared towards how these models interact with the "real" world.
It seems to me that we are on the fast ramp to making the paperclip maximizer a reality. imo the most likely "dystopian" future that we'll get to see. These models have no concept of their RL env sandboxes and the real world. Extremely interesting to see how they compete in trying to get finite resources and work with constrained supply chains.
I feel like the two companion pages - AI agents explained [0] and Agentic AI examples [1] - are kinda more telling than the Muse personal agent itself? I don't quite get why we have these two posts at all. A personal agent is more show than tell. Live demos with an actual agent doing day to day things would have been the ideal thing to show off. A version of the infamous Apple Intelligence ad but live.
At any rate, it would be interesting to see what the actual capabilities would be. The Meta AI integration with WhatsApp actually sucks so bad. Hope it's better than that.
This reminds of when I was in school and we had 3 categories of notebooks.
1. A "rough" notebook to jot down notes
2. A "classwork" notebook to present those notes properly to the teacher and get them corrected and
3. A "homework" notebook to finish homework assignments that used to be given
It's quite a discipline to jot things down in the rough work book and then write out a "neat" version in the class work book.
I stopped doing it once I finished school and realized I didn't really need to do it anymore once I understood what I was writing out. But maybe it’s a habit worth returning to.
https://github.com/DeadMeme5441/arrodes
It seems to be a bit more integrated than what you describe here because it's completely built around a Clojure REPL