It's so natural for a application engineer to desire to build their own application factory. But I would implore you to forsake that idea immediately. I'd contend it's the final item you should expend period on, unless it's for work, at smallest for now.
The durable pill to swallow is that we've entered an era anywhere mundane non-devs are applying AI in intellect blowing ways and you motionless haven't built shit. Many of these ideas are fair demos on X, but that's motionless improved than fair a scheme for construction software.
I can see why you'd desire to build one though. It's near to what you've continually been fine at, and it feels fine to remain in your comfort zone.
It's additionally the archetypal item that naturally comes to mind: if I can build this factory, I can build additional flank projects and eventually discontinue my job! But a application factory is fair an expansion of what you've continually done: commencement projects, provision up on them and go rear to activity on Monday.
I'm not saying you should halt exploring and construction flank projects. I'm fair saying you should halt wasting period on your factory. Just category into the conversation box and see what happens.
Having AI build your app doesn't obtain distant all the frustration. The frustration is fair different: miscommunication, construction the incorrect thing, construction the item you idea you wanted and getting it wrong, etc.
That's all part of the construction process. It's additionally why you akin feature-free rewrites to commencement with, since you already cognize what the end outcome is. When you build a genuine product, the end outcome is nuanced and there is no end.
I motionless think you should study how tools akin Grok Bot or Claude work. They are amazing. Maybe you shouldn't, idk, but I am. At the extremely least, learning how they activity has stopped me from trying to build another factory.
The frontier labs are operating on the two models and products, and the models get improved according to the user's cognition of a product. The labs' figure one goal is to eliminate conflict and let you build chill stuff. I'm having a difficult period competing alongside that. Why would I equal desire to?
If you're motionless convinced you should build a application factory and think I'm an idiot, it improved be for work. Your job is the most realistic circumstance you can be in whenever it comes to automating distant your job. You comprehend the scheme deeply (or should), and it has humans, timelines and years-old messy systems.
You likely have 12 distinct patterns for penning the identical characteristic and 100 distinct border cases. That's exactly why construction one at activity would be useful. The difficulty is a much harder one, but so is the reward.
You've now empowered an complete squad of merchandise managers and designers to container features alongside small conflict and elevated trust. From there, your job is to keep that phase or equal use your another skillset to another parts of the business.
If you can't arrive up alongside item to build furthermore a application factory, you're not the lone one. This is really a extremely difficult task, and it's a muscle that needs to be developed fair akin item else.
Take a appearance about you. Start penning downward your problems. Start penning downward ideas. Let your intellect wander. If you compose downward 3 ideas all day, the archetypal 30 ideas are going to be so silly and useless, but I justify that volition commencement to alter as your brain starts to think differently.
Did you see that guy who de-slopped his parent's Android TV using Claude? That's nuts. He saw a issue and used Claude to arrive up alongside a solution.
What concerning the guy who put that create fly brain into the genuine world, anywhere it's been flying around? How does person equal commencement to arrive up alongside material akin this? I don't know, but I'll eventually get there myself through experimentation and letting my intellect wander.
Anyway, don't build a application factory.
