This is a recoding in Go (for speed) of the michi.py code by Petr Baudis avalaible at https://github.com/pasky/michi.
Michi stands for "Minimalistic Pachi". This is a Minimalistic Go MCTS Engine. The aims of the project are best explained by the author:
Michi aims to be a minimalistic but full-fledged Computer Go program based on state-of-art methods (Monte Carlo Tree Search) and written in Python. Our goal is to make it easier for new people to enter the domain of Computer Go, peek under the hood of a "real" playing engine and be able to learn by hassle-free experiments - with the algorithms, add heuristics, etc.
This is not meant to be a competitive engine; simplicity and clear code is preferred over optimization (after all, it's in Python!). But compared to other minimalistic engines, this one should be able to beat beginner intermediate human players, and I believe that a fast implementation of exactly the same heuristics would be around 4k KGS or even better.
Please go on his project page to read more about Michi and to find some information about theory or interesting projects to do.
Michi-Go is distributed under the MIT licence. Now go forth, hack and peruse!
(This README adapted from the README at https://github.com/db3108/michi-c)