Alex Rogozhnikov, a machine learning researcher, writes explanatory posts with interactive visualizations on topics like Hamiltonian Monte Carlo and gradient boosting.
HN commenters point to his MCMC and Hamiltonian Monte Carlo post as a nice explanation with good visualizations, and to his Python and Fortran binding post.
Why not using both ? You can reuse existing fortran code withing Python, and benefit from the prior experience and speed while enjoying the qualities of Python.
There are several ways to bind Python and fortran together. E.G http://arogozhnikov.github.io/2015/11/29/using-fortran-from-python.html
Good paper indeed. And here's a nice blog post with nice visualizations for MCMC and Hamiltonian MC.
https://arogozhnikov.github.io/2016/12/19/markov_chain_monte_carlo.html
described here:
http://arogozhnikov.github.io/2013/07/20/blog-post.html