Nice app! Feedback since you asked: The most obvious must-have feature IMO is to paste automatically. Don't require me to hit a shortcut (or at least make it configurable)
The next most critical thing I think is speed and in my tests it's just a little bit slower than other solutions. That matters a lot when it comes to these tools.
The third thing, more of a nice to have is controlling formatting. By this I mean - say a few sentences, then "new line" and the model interprets "new line" as formatting, not as literal text.
Nice looking site and design - kudos. I don't think there is anything wrong with the idea but you do have to compete with easier alternatives (e.g. credit cards, gofundme, craigslist). You can't just be different than the alternatives to effectively compete, you have to be better - especially for the creator who in your scenario needs money to make ends meet.
As others have mentioned your biggest challenge is the two-sided marketplace, it is unfortunately just about the hardest thing to pick for a startup. You won't get creators until there is a large enough volume of supporters for it to be worth it to them (why set up a campaign if no one responds). You won't get supporters until you have enough creators to match services they might want.
Many marketplaces like this start with at least one of two things - they either already control one side of the market via relationships (they have access to a large amount of supply, for example), or they have a lot of capital they can use to fund the concept until it gains traction (by funding creators for example). Ideally you have both.
My advice, given the time you have already invested, is to figure out how you can niche down very narrowly to start. Pick one geographic area, and perhaps one service type, and then grind it out manually to match supply to demand. Maybe you're matching creators who tutor on the side, to students in Seattle (or whatever). You want to be narrow because it's too hard to build supply and demand if you're broad, especially by yourself without funding. You'll have to try lots of hacks to see what works to get initial users - it will be manual, you will have to talk to a lot of people, it won't "scale" but that is ok to start. If it works you can grow into other markets later (example: Facebook started only for Harvard, Uber was only black limos in SF).
Also consider looking for a charitable benefactor who will fund creators (even if they don't use the services) to get a bit of energy into the marketplace. If you can find someone, or a company that wants to do this for a bit of good will and press you'll have some capital to fund the initial creators.
It solved Day 3, Part 1 for me almost instantly. It used the sample strings as opposed to the full input file, which makes sense because I didn't give it the full input. Modifying the generated code to handle the entire problem is low effort to extend yourself, so I was quite impressed.
Regarding your question - my current guess is that you probably are going to need smart people to tell ChatGPT what to do and check that the output is sane. I expect this is where quite a bit of tech is going - at least most of the basic scaffolding.
The next most critical thing I think is speed and in my tests it's just a little bit slower than other solutions. That matters a lot when it comes to these tools.
The third thing, more of a nice to have is controlling formatting. By this I mean - say a few sentences, then "new line" and the model interprets "new line" as formatting, not as literal text.