I've worked on memory and performance optimizations on algorithms written in Java by scientists, and I'd definitely agree - code review is really needed. Some t

I've worked on memory and performance optimizations on algorithms written in Java by scientists, and I'd definitely agree - code review is really needed. Some things are better handled while the design of the algorithm is taking place.