Is there a way of reverse engineer a web server programming language based on how Secured Random results?

I want to know if you can determine the programming language of a web server based on the random outcomes, like UUID's and random patterns created by different kinds of pseudo random generators.

For example, a standard random UUID contains 122 unique bits and is generated by a simple predictable LCG. However, a Java random seed only contains 48 bits. This means that there are not enough seeds to be able to generate every UUID. Not every UUID is possible to have been generated by the LCG.

Using that, to what extend is it feasible to do that. Most websites will probably use Secure Random for generating values, would you be able to apply the same logic as up above?

Note: I know that it is not possible to crack the seed, but maybe there are still patterns.