"Quant circles" = sector of finance where nerds and scientists cook up programs and algorithms to rip off the market...err...I mean make "wise investment decisions" by "mitigating risk" through the use of data analysis in Python. I've been using Excel essentially, but want to do more complicated math and filter data more efficiently. Programming in VBA is not bad, but still clunky feeling to me.
while trying to find a good example video, I stumbled onto a vid from the guy who wrote this book :
I want to use it as a way to make prototype programs like ,
- finding the angular frequency vs k-space data for a certain input of initial conditions and printing it on a nice graph of the 1st Brillouin Zone (I think the matplotlib is for that),
- control my arduino for prototyping (a possible problem with Python 3.4, seems to need 2.7), and
- tinker around with analyzing & displaying stock info.
I selected Python based on input from engineers I've discussed it with. But they don't smoke weed, so they aren't on here. Physicists around me use it, too. I'm sort of forced to learn Java on the side through programming the Arduino, as it is, in the native interface. But I mostly hack shit on there, and learn what I need when needed. I am not a "coder" in the proficient sense, at least not yet. 
I will be taking a course in the summer either on Python or Java, but I am thinking I should probably take the Java course based on what I'm hearing from you and Doer, among others.
If it can handle fractals and Fibonacci, it can handle the math I'm interested in.
In this book I'm trying to learn from, it opens up with a Black-Scholes model in Python. So that's already some funky math incorporating the libraries I mentioned earlier (Scipy, mostly). I take it you never did any weird calculus with it?
I'm quite familiar with Black-Scholes, I was a Finance double major in college.
So, to this point, if you are only looking to do non-learning scripts, I think Python and Java are a bit of overkill. If you are using this to build models and test theories, you don't need multi threaded application speeds.
BUT, with that being said, you are tinkering with Arduino boards, which I then can assume you are tinkering with Android and/or some flavor of Linux. So this stuff isn't totally out there for you. So I'd say go for it if you like Python..
I've never done any weird math functions like Black-Scholes theorem, but I've used it to calculate associations between different indices of text with advance AI algorithms.
Now, with that being said, Python is not really a prototyping language like PHP or Ruby, or even Javascript (with Nodejs), unless you use libraries like Django. Now, if you have a set of math libraries that you would like to take advantage of, and they seem to only be in Python, well then, there is your answer. If not, and you think you might write your own advanced equations, then any language would do, including Javascript (with Nodejs). Why? Because it is waaaaay faster, in terms of scaffolding.
