Not sure if this counts. I utilise Prima (a numerical optimisation package developed by a Chinese prof, based on algorithms from Cambridge) in my research on nuclear physics :
No, it is not difficult. It is way more than that. It is frustrating and depressing according to my experience attempting to modernize a FORTRAN 77 codebase (I gave up finally). Navigating through a quagmire of GOTO statements felt like a desperate endeavor. It's hard to fathom how someone could work on such a project for three years, especially if it appears to be a solo effort.
Upon examining the original FORTRAN code and accompanying papers, I realized that the complexity of the task was far from trivial. Working alone for three years, especially on such a complex codebase, requires a significant amount of dedication and perseverance.
This is truly impressive! While I'm not well-versed in these specific techniques, I do recall learning about Powell during a mathematics course in the past. It's reassuring to know that capable individuals are maintaining his work.
Some time ago, I attempted to update an antiquated FORTRAN program for a particular project, but regrettably, I had to abandon the effort. Navigating the labyrinth of countless GOTO statements was both challenging and disheartening. I became genuinely concerned about the potential impact on my mental well-being.
That being said, I have a profound admiration for the individual who successfully modernized such a complex FORTRAN project. Their achievement is undoubtedly commendable!
https://github.com/libprima/prima
Not something as big as tencent, alibba, ..., yet useful enough for my group !