Think about what interests you. What gets you excited. What would keep you up at night, thinking about the possibilities. What would make you get out of bed at 3am to go try some idea that just popped into your head.

Then find open source projects in those spaces, or in related spaces. Find out what their contribution policies are. Get involved in their community. Get advice on low-hanging fruit that you could take a crack at, but which would also help address current pain points.

Or, conversely, ask yourself what kinds of things about your city/town, neighborhood, or house that concern you.

Do you live near woodland areas? Near the Woodland-Urban Interface (WUI)? Are you concerned about wildfires in your area? Are there local groups you can join of like-minded people? Do they have issues that you can help them solve? Could you help them run their web server for their organization? Could you help them do Structure Ignition Zone Evaluations? Have you talked to the IBHS about making your own house a Wildfire Prepared Home?

Or, are you talking about building physical stuff? Is there a local group in your area that does blacksmithing? Or Habitat for Humanity?

There are an infinite variety of options here. But I think you have to start with what interests you, or what concerns you. Then follow the leads from there.

The basic principle of management is "finish what you start", see

https://en.wikipedia.org/wiki/Kanban

and

https://en.wikipedia.org/wiki/CONWIP

To take an example the people at the cafe in the next building over have a certain capacity to make food including heating in the over and all that. They don't take your order until the WIP drops below their threshold so they maintain their workflow without being overwhelmed.

Most of the time I have three "side projects" and I am really making progress on two of them and the third is aspirational... and that's OK! What's not OK is when you are starting new things over and over again and never finishing them.

stuff? just start building

for more specific projects you may have to plan first

WHY you want to build stuff?

Figure out answer(s) to that question, and chances are the how, what paths / projects to pursue, will follow from that.

Find broken stuff to take apart, and see how it works. You'll build up a set of tools doing so. Over time you can learn to put them back together again, and eventually to repair things.

This works for mechanical stuff, electronics, and even software.

Also, you probably can't go wrong watching Dan Gelbart's 18 part YouTubee series about prototyping things:

https://www.youtube.com/playlist?list=PLSGA1wWSdWaTXNhz_YkoP...

For the same reasons you start dancing by dancing, you start building by building.

Building is easiest when building is ordinary.

The only way to make it ordinary is building when you think about building.

Otherwise not building remains the default state.