Concept: a 3D printed board serves as a structure for holding components so they are easy to solder. This can be described as dead-bug soldering with support structures. Note that Dead-Bug Soldering is useful for high frequency circuits - by the way.

3D print a circuit board with channels and physical infrastructure for holding all components.
Bend the leads so they touch, and use wires to solder as needed. For through-hole components, this is fine.
This is not fine for small feature advanced microelectronics, but 2.54 mm components would work.
Soldering is quick for 2 wire components that touch each other. We can have the 3D print serve as a holding structure that makes soldering easy.
2nd option: wires snap in place if printed from elastomer. Ga 10 or 8 solid copper - just snap it in. Locking, snap-in Circuit cover finishes circuit. Components become 100% reusable!
3rd connection option: Not also - if 2 leads are inserted into a hole, the hole can be plugged with a solid copper conductor for good contact.
Terminal blocks can be 3d printed in place!
Note that we can make connectable modules with this technique, where 3D circuit blocks can plug into one another.

Other ideas

TEST using a dual extruder printer to place conductive fulfillment in a print.
- such as, but would want lower resistance than 100kohm... https://hobbyking.com/en_us/conduct-electricity-abs-3d-printer-filament.html?countrycode=US&gclid=CjwKCAjwwZrmBRA7EiwA4iMzBI_jQE3oiTDjLxUGxKI7EA9xk4DzdBZ3MGNXzdm0TRXLhxfncVG2HBoC0qgQAvD_BwE&gclsrc=aw.ds
TEST: Conductive filament could be shaped in such a way as to achieve a spring like grabbing of electronic parts?
TEST: Could other features be printed to push SMA or other topologies onto these printed pins? think of placing the parts and having pressure applied from a heatsink above to press onto these special "pins"

Pros:

Full 3D, body-integrated circuits can be produced.

Cons:

Cons: limited to relatively large components - such as atmega328p DIP. But socketing can extend range of applicability, such as SOIC to SMT [1]
- Could this be mitigated by printing a nest for SMA parts then using conductive ink where they part is depressed onto the cradle?