Hi adrenaline,
In mechanical engineering, the efficiency of a pump or compressor is also known as the isentropic efficiency which is equal to the amount of energy an isentropic process would require to compress or pump some fluid, divided by the actual amount of work required. (isentropic means the process has no
change in entropy)
Similarly, the efficiency of an expander is also known as isentropic efficiency which is equal to the amount of energy or work that is removed from a fluid by an expander, divided by the amount of work removed if the process were isentropic.
Here's a reference that may help:
http://www.taftan.com/thermodynamics/ISENEFF.HTM
There are other types of efficiency in engineering, but I think these are the most applicable to what you're asking. They apply to a mechanism and its thermodynamic efficiency.
The point is that in order to calculate efficiency, you need a definition. There is nothing intrinsic about the two definitions given above, they are simply definitions that we can all use and makes the most sense. It doesn't mean there really is such a thing as an absolute efficiency. For example, an isothermal compression process will require less work than an isentropic one, so such a compressor would actually have an isentropic efficiency greater than 100%! If a compressor ended up putting a lower pressure out than the pressure in, the isentropic efficiency would be less than 0%. It all depends on how you define efficiency.
So to answer your question, I think it doesn't make sense to ask what the efficiency of a person doing a pushup is. One needs to define efficiency in order to measure it and give it any quantifyable value. Depending on how you do that, efficiency can be anything if the process isn't somehow intrinsic to physics.
Side note: some processes are intrinsic, such as the overall increase in entropy in the universe given any process, but I don't know that really applies here.