Building your own gear for the workshop (be it electronic or mechanical) is (IMHO) one of the most rewarding things you can do.
The great thing about having enough tools and enough knowledge is that you can then extend that basic tool/knowledge-set almost without limit by building on what you have.
My engineering workshop has many cool tools that I've built myself, including a metal-spinning lathe, pipe-bender, spot-welder and a myriad of other 'special purpose" additions that I've built because I needed them and they made a great project that taught me new skills while also saving me money.
The same goes in the lab.
I remember when I was an impoverished small business owner working on RF gear (mainly CB radios) back in the 1970s and all I could afford was a 5MHz scope. I simply built a beat-oscillator and mixer to effectively extend the useful bandwidth of that scope to the 27MHz I needed to get a picture of the modulation envelope of those CB radios.
I've also built countless bits of test-gear over the years and each one was a fun project.
Just make sure you get some good basic equipment to start with (scope, DMM, etc) from which you can start your building process and against which you can calibrate your new creations.
As for what you ought to be building -- that entirely depends on what you plan to do in your lab. The needs of an RF engineer/experimenter tend to be different to those of a digital engineer/experimenter so different tools have different values to each.