Since working a student job at SINTEF Ocean in Trondheim, Norway, Christian(cnheider) has been developing and supporting software suite named Neodroid.
This suite comprises tools for prototyping reinforcement learning environments codenamed 'droid', establishing connection between a running simulation and client machines for collection of observations and interacting with the enviroments codenamed 'neo'. Further more a whole host agent-centric implementations of State Of The Art reinforcement learning algorithms is provided under the codename 'agent'. Example environments are continually added and hosted at 'Examples'.

Working daily with Computer Vision, Christian develops novel techniques and provides implementations at 'vision'.

All of the mentioned tools are still being developed and improved and are hosted at pypi, along with long range of convenient python and framework extensions listed below.

• apppath
• warg
• draugr
• heimdallr
• munin
• A much much much more and expanding.

With your sponsorship, I will able allocate more time into bringing you more simulation and game-making tools with a focus on machine learning through reinforcement.