In finance it looks to be all but dead. 10 years ago, every fund I knew used Octave, or more likely Matlab. Now I don't know of a single peer that uses Ocatav

In finance it looks to be all but dead. 10 years ago, every fund I knew used Octave, or more likely Matlab. Now I don't know of a single peer that uses Ocatave, and only 2 that use Matlab. There are simply better alternatives out there. Matlab and Octave are too "standalone" and too "slow" to implement into any real-time trading model. I saw two big use cases for them 10 years ago, 1: writing Neural networks from scratch where you culd apply vectorized operations over multiple data at the same time. 2: Engineering work as Matlab had a huge, and just as importantly, well tested set of fortran functions for almost any engineering discipline. Both of those uses cases were supplanted by other tools. The former by tensorflow, pytorch, pandas and numpy. The later by excel in alot of instances but also the python eco system of numpy and bespoke libraries on top