Postdocs are really weird and random. I mean, I probed a little and think it might have been a little hard to get a quantum computing one circa 1998 -- it might have been a little early but I think it would have been easy to establish a career in that if you got at the right time. [1] I wound up getting one where I had a lot of latitude to work on "complex systems" and "quantum chaos" and wound up accomplishing little at a brand new institute in Germany which had a lot of startup funding.

About ten years later I got offered (but didn't take) a postdoc at one of the most prestigious centers for the semantic web on the strength of a good, if popular, talk and a single conversation and I wasn't even looking.

That is, if you find the right personal connection you can get a postdoc without having to have the right formal qualifications or run the gauntlet of a selection process. If your viewpoint is a stable and safe academic career you might be just setting yourself for exploitation but if you have a passion for something and are looking for a fulcrum with which you can move you world that might be your path.

[1] ... good timing may be the royal road to academic careers

What sort of equipment? I've seen surprising stuff put up on ebay, but you still eat some pretty big costs + freight shipping.

Otherwise it'll be tough. I've wanted some fancy niche Rhodes and Schwartz products in the past, but I had to come to the realization that it's probably not accessible to me.

I learned everything I know about CNC machines and plasma cutting and 3D printing from local hackerspaces and makerspaces.

If your idea/invention can actually make people money, then VC funding would be an option. Either that or you need to bootstrap the startup costs.

Also, building subscale POCs could be a good option depending on your idea and help cut down costs significantly, especially if you can use or modify COTS parts to make it work for the POC.

For example, when Rocket Labs was starting out they salvaged components from junk yards and so on to keep costs down like paying $5 for a used valve and refurbishing it rather than pay $100 for a new one (Peter has talked about this in various interviews).

Write a technical paper. Email college professors, see if you can pay them as consultants to review the paper.

Once you have reviews and other people say its plausible, you can go present to VCs and acquire the funding.