A correct computer program and a sound argument piece are the same
All that software does is interface between lower-level building blocks and higher-level desired goals and behavior.
A program, when executing through a microprocessor, is yielding that silicon mechanism through a discrete sequence of physical pheonema with well-understood state outcomes.
More concretely, say a processor has three registers, A, B and C, which are physical silicon capable of holding, while powered, say, an integer number. An ADD instruction will “turn” the processor's mechanism that sums A and B, then store the result at C.
Whether the program will achieve any goal by doing that is up to the programmer and the user, but that instruction will be correctly followed.
Arguments are much fuzzier than that, but they can still be understood in separate pieces.
Say I argue that, since I saw evidence of A and I think A often leads to B, B is true.
Breaking up the piece, my thesis is “B is true”. That's my claim with this argument.
First, I am also claiming that “I saw evidence of A”. That can also be questioned by, for example, ensuring that I did see it first-hand, how did I experience it or, for physical quantities, how did I measure it. Alternatively, maybe I have reason, incentives or biases to misrepresent A. All of these can be examined and verified in isolation from the rest of the argument.
Then, I'm claiming that “A often leads to B”. That's a judgment call that you can agree or disagree with based on your own experience or reason. Maybe I didn't take into account an additional factor C that cancels A. Anyways, that claim also can be examined in isolation from whether A is true or not. “Let's assume that...” is often the rethorical device that allows considering the outcomes of A without discussing the merit of A.
Finally, the conclusion or thesis, “B is true”. If you are willing to believe that “A is true”, and you agree with me that “A leads to B”, the only logical conclusion is that “B is true”. You can attack the evidence and the argument, but once both are settled, the conclusion is what it is.