Has anyone seen success with them? I feel its better to put something out thn nothing at all.
Spend as little as two weeks to build the prototype or MVP, and start showing it to potential users/customers. If it solves a real problem for them, they use it daily and commit to pay for it, then you can think about enhancing the no-code app or moving to a hybrid or all-code approach.
Test the idea first with least time and effort, which is surely possible using no-code builders.
Don't worry too much about the stack. The basic principle is you'll have to iterate eventually, and so whatever you build will be scrapped once the users touch it anyway.
If you're actually planning to build anything remotely complex (like something that offers a service that users will directly interact with), then they're probably not great solutions, especially for anything past the initial trial stage where you're doing everything manually to gauge interest.