\section*{Acknowledgements}

I would like to first express my gratitude to Professor Nicola Serra for letting me do this thesis in his group and being very supportive and trusting overall.

A huge thanks goes to Dr. Albert Puig Navarro, who was not only a great supervisor but also a core member of the \zfit{} project, a co-author, code reviewer, advertiser as well as main author of \zphasespace{} and \zamp{}, both libraries that are very closely interconnected with \zfit{}. Without all the discussions about the large and small details of the library and use cases, the pure coding and reviewing contributions, this project would not be on a comparable level and may never grew over a small collection of scripts.

I would like to greatly thank Dr. Rafael Silva Coutinho who is also part of the \zfit{} core team and part of bringing the project to life. His contributions in discussions with a more user sided view as well as extensive usage with real use cases is, next to code and documentation contributions, a great support in designing and creating the library.

There are a lot of members in my group that I would like to thank as well for a variety of things. Be it for discussions about the development of \zfit{} and code snippets I like to specially thank Dr. Abhijit Mathad, Dr. Julian Garcias and Dr. Oliver Lantwin. For the usage and trial of the library including new features where I like to thank Davide Lancierini, Sascha Liechti and Martina Ferrillo. And last but not least for the idea of the library, the technical and user sided experience with an already existing tool and helpful advices I would like to thank Dr. Andrea Mauri and Michele Atzeni.

A specially thank goes to Prof. Anton Poluektov for several discussions and 
whose library \texttt{TensorFlow Analysis} was a major inspiration to build 
\zfit{}.

Furthermore, I would also like to thank the scikit-hep community for various design discussions, especially Dr. Eduardo Rodrigues and Chris Burr.

A great thanks goes to the University of Zurich and CERN for offering the 
opportunities to do this kind of research by providing the adequate resources.

Many thanks also go to Google and the TensorFlow 
development team for open-sourcing the library and thereby allow libraries like 
\zfit{} to be built.

Last but not least I'd like to thank two persons not only professionally for 
their support but also privately for various things, too many as they could 
even remotely be summarised here: 
Simone Steinbrüchel and Patrik Eschle