A Mu-Zero based game environment that consumes DASK tasks (ML tasks) and produces performance metrics

I am looking for a freelancer who can develop a Mu-Zero based game environment that can consume DASK tasks (ML tasks) and can produce performance metrics. The platform should be written in Python and specifically provide performance metric for the speed of the game. I will provide the machine learning task and be specifically focused on the MuZero model. This platform must take into account the parameters found within the DASK tasks as well as the levels of performance it must expect. The expected output should include information about the accuracy, speed, and error rate of the game environment. I look forward to working with a professional developer who can bring this project to life.