I've developed, maintained, and extended web platforms that all could benefit from what Pegasus seems to do. These platforms have driven up into the million-per-month range of sales for consumer (and team!) based products. The vast majority of the platforms are trash, yet power jaw-dropping amounts of sales and usage. I think it perfectly reasonable that Pegasus has all the underpinnings these kind of platforms would…
The most concise answer as to why I haven't done it is that I'm just one person, and I'm not smart enough to have figured out how to do it yet in the ~2 years I've been working on Pegasus.
To me the huge difference between a code starter kit like Pegasus, versus, e.g. Shopify, Webflow, or any other hosted no/low-code product is that at the end of the day you end up with a codebase that is maintained by developers in a totally standard framework and language that they can do anything with. No code / low code tools are inherently difficult because they have to put in place this firewall/abstraction layer that hinders your ability to customize. The same thing that makes them great for simple things makes them bad for difficult things.
So anyway, I think a hosted version of Pegasus should still give the power of anything you want to do with a Python/Django application, and I'm sure that's possible to do in a hosted way, but I find the prospect of building such a thing in a scalable, secure way quite intimidating.