The result is a simple web app that generates short, personalized bedtime stories and narrates them in my own voice. The codebase is intentionally rough in places, but the app works and has been usable in a real setting.
What surprised me most was how easy it was to iterate on a real, deployed system without getting stuck on early architectural decisions.
Stack: FastAPI, Supabase, Fly.io, Vercel Tools: Cursor, Lovable, Claude, ChatGPT
Curious how others here are approaching AI-assisted development, and where you draw the line before hardening things for production.
Link: https://sleepli.app
dangus•1w ago
I think I would be rather hesitant to use voice cloning to read to my kids before such things were studied by medical professionals.
I would also be hesitant to use this because I’m not a lazy piece of shit.
Another stupid AI app to file under “Nobody asked for this.” Do we really need AI-personalized bedtime stories?
Have you tried using your brain as a parent? You already know your child, make up a story with them. You can literally just do it on the fly because you have a real brain with intelligence that isn’t artificial.
But who knows maybe your ideal customer profile is people with severe lack of judgment.
(Call my comment harsh if you want but I don’t think it’s harsh enough. This is like “downfall of enlightened society” type of shit.)
ealpopa•1w ago
dangus•1w ago
Your child wants to hear a story with grandma’s voice, then all of a sudden later they’ll be talking to real grandma as if she was really saying the things in the story. Grandma will have no recollection that she said any of those things because she wasn’t there and didn’t say them. Creepy stuff.
Psychologically I can’t see how introducing this element to a child could be a good thing.
Good point about story books. Anyone who wants to use this should just buy a quality book at their local bookstore or check one out at the library.
No one asked for this dystopian AI slop.