Basic physics principles combined with the above. Being able to not only see how something works, but to be able to at least estimate it's parameters is invaluable to designing anything new
Same for me. The book and its tools are extremely empowering. In general, anything from Sanjoy Mahajan (the author) is extremely insightful:
-Street Fighting Mathematics
-The Art of Insight in Science and Engineering
-Lecture notes on signals & systems (a bit harder to find)
-Lecture notes on thermodynamics (real-world insights rather than endless manipulations of weird partial derivatives)
Afaik he is currently working on a textbook on mechanics.
I think skills like these are much overlooked in academia. Some of the most engaging and rewarding work I've had in my 32 years in computing professionally has been spent better understanding the problem space and the limits to the signals available therefrom. Knowing how to better map the problem to the solution space deftly interweaves the art of science with the craft of engineering, greatly enriching the journey betwixt.
It's a set of course notes from MIT 6.055 - "The Art of Approximation in Science and Engineering". They have been edited and updated into a book as well:
http://web.mit.edu/6.055/book/book-draft.pdf
Basic physics principles combined with the above. Being able to not only see how something works, but to be able to at least estimate it's parameters is invaluable to designing anything new