I'm not a rocket scientist (heh) but I'm imagining a many-stage nuclear rocket. You'd have a whole bunch of acceleration stages followed by a much smaller numbe

I'm not a rocket scientist (heh) but I'm imagining a many-stage nuclear rocket. You'd have a whole bunch of acceleration stages followed by a much smaller number of deceleration stages followed by the payload. You wouldn't need as many deceleration stages since you've already spent all the mass in your acceleration stages. Building such a thing in space would involve sending up the stages themselves with some kind of heavy lift vehicle, assembling them in orbit, and firing it. The number of stages you could assemble would be quite large since there's no air resistance or (relative) gravitational stress in space. You'd just have to not exceed the tensile strength of the material during acceleration. You could put together 100 stages or more I'm sure. It's one rather brute force but workable way of reaching some nontrivial fraction of the speed of light. Someone should do the math on this for fission (NERVA-like), thermonuclear pulse (Orion), and pure fusion.