I had a booth about this at the Bay Area Maker Faire lately.

If we're all printing the same object on our 3D printers, it's proooobably a lot less trouble to just have someone injection mold it and save us all the trouble. 3D printers are really great for one-offs and mass-customization and things like that. Aaaaaand, I feel like it's kind of an under-appreciated problem in 3D printing. Because, yeah, CAD is hard and we're never going to reach a world where every 3D printer owner is very very comfortable with CAD, and so it should be more of a concrete goal for the 3D printing community to make sure that we're focusing on this problem. It's important that every 3D printer owner can do at least some amount of tweaking and customizing, otherwise we're failing as a community.

Now, I don't Tinkshame. I spent a lot of time learning Blender, FreeCAD, and OpenSCAD to prove Naomi Wu's assertion that we should all just get over ourselves and use TinkerCAD. The only real problem with it is that it's not really free, it's "free at the pleasure of AutoDesk" where they could raise the "Mission Accomplished" banner at some point and turn it off. And there's not really an open source version of it for roughly the same reason that random thingiverse models are always kinda halfassed and bad. Doing a good TinkerCAD-but-actually-free-by-some-definition is actual work to get everything right and polished and documented and bug-free and nobody really wants to pay for it.

Also, maybe I am pedantic and obsessive, but I don't really like screwing around too heavily with models in a slicer, so I'd rather they take some of the magical code in the OrcaSlicer/PrusaSlicer/SuperSlicer tree and actually organize it into something that could be TinkerCAD-esque?

Anyway, the core of the talk of my booth was systems and libraries of 3D printable objects. So, for example, there's the Honeycomb Storage Wall system and some of us have been writing some neat lil OpenSCAD libraries and models for it (and another group of people have been doing similar things in Fusion) where you can make a parametric model so you can measure your flashlight and print a cute 40mm holder for it based on the measurement without having to model things from scratch and it'll click into the HSW wall and it's fine unless you are married to someone who has ommetaphobia and then you need to make sure that the honeycomb is the same color as the wall. And the same is true for Gridfinity, just you can put that in the drawer.

And there's also a lot of parametric models. I'm not sure what you are looking to print, but there's a decent selection of people who have done stuff in Fusion or FreeCAD or OpenSCAD where you can download the model and change the parameters to get it a lot closer to what you want without going through all of the drama of making it all over again.

I love using OpenSCAD. I've got a buncha years of experience using various 3D modelling tools at various times and so I can use Blender or FreeCAD quite well actually, but in the end, I do a lot of functional bits and it's so darn easy to just write some code because, actually, I've been working as a professional software engineer for quite some time.

So... dono, it depends on your aspirations? There were a good number of Gridfinity-like systems that were around before Gridfinity came out and they were ... ok, but not great, but then Gridfinity came along and did a boxy-box system just like was already there but with some interesting tweaks and making it more amenable to real customization and suddenly everybody went gonzo over Gridfinity in particular. So you might not be just making a thing that exists in a dozen forms better if you borrow an idea and make your version of it.

Also, I learned 3D modelling tools mumble mumble years ago in a failed attempt and/or dodged-bullet because I'd wanted to do games or special effects as a kid. The software I learned on is long gone, but it turns out that once you are thinking about things, it tends to stick? Which means that I learned pottery while visualizing the objects I was making on the wheel as if they were in the CAD window of my mind, got good at photographic lighting based on what I'd observed in the 3D program, and then transitioned back to CAD because I wanted to make things, so it's kinda one of those things where you probably won't waste the time spent.

tl;dr: I learned OpenSCAD, FreeCAD, and Blender to prove that Naomi Wu is right and we should all get over ourselves and use TinkerCAD and ... she's still probably right, LOL.