I made one for myself (and my son, both type 1 diabetics) and decided that I might roll it into a possible possible product:
https://carbcount.ai
Allows diabetics to easily get carb estimates for their meals by just snapping a picture. Free for now, would love some feedback.
I built the basic idea in like 2 days using Google Antigravity. It's all serverless on Google cloud run and uses Google for auth, doesn't store any data. MVP while I am working on the marketing and testing.
I used a similar concept I originally used to train my son (9yo) to be better at counting his carbs with a custom GPT. But that idea didn't work for others to use. Most people don't know what a 'custom GPT' is, so explaining it was difficult. Web-apps everyone gets, it has PWA support so you can save it as an 'app' on your phone.
I am a software engineer so it's already in my wheelhouse and it just costs me the tokens right now so it's cheap to run.
Would love feedback or questions if anyone wants to know more.
All I am saying is that if you are expecting it to fail for you, that is absolutely what it will do. You weren't good at typing on your first day either. I went from a decent engineer with 20 years experience to a 10x engineer able to take on any problem, in less than a year. All because I learned how to use the tool effectively.
Just dont give up, get back on that bike and keep peddling. I promise it will amaze you if you give it a chance.
This is a tool, just like all the other ones you have learned, but it will make you a far better engineer. It can fill all those gaps in your understanding of code. You can ask it all those questions you are unwilling to ask your colleagues, because you think you will sound dumb for not knowing. You can ask it to explain everything again if you still dont get it. It is powerful if you know how to use it.
To all those devs saying "i tried it and it wasnt perfect the first time, so I gave up", I am reminded of something my father used to say:
"A bad carpenter blames his tools"
So AI not going to answer your question right on its first attempt in many cases. It is forced to make a lot of assumptions based on the limited info you gave it, some of those may not match your individual case. Learn to prompt better and it will work better for you. It is a skill, just like everything else in life.
Imagine going into a job today and saying "i tried google but it didnt give me what I was looking for as the first result, so I dont use google anymore". I just wouldnt hire a dev that couldnt learn to use AI as a tool to get there job done 10x faster. If that is your attitude, 2026 might really be a wake-up call for your new life.
You might want to think about using cursor instead of attempting to use a web interface to create full app. Cursor can do better targeted changes and allows you to tell it individual file to look at for changes or the whole codebase.