The aim of this project is to design and implement a photonics realization of a liquid state machine, with the potential for versatile and fast signal handling. Optical information exchange and processing are major challenges in nowadays and future photonics networks. Huge amounts of data need to be handled/processed in photonics, but also in other fields, like biological or economical systems. Despite of increasing computing power, new approaches and procedures in photonics are desired and required. Novel promising concepts utilizing complex dynamics are being developed, one of which is reservoir computing. Reservoir computing represents an alternative approach towards computation. It is based on mapping a dynamical input state onto a high-dimensional state space. This is realized by feeding a dynamical system called reservoir with an input stream. By this it is converted into spatiotemporally distributed states of the reservoir. Specific computations are then realized by a mapping of these spatiotemporally distributed states on output states based on a trained readout. It has been shown that this kind of reservoir computing serves universal computational properties, such that any potential operation can theoretically be realized. Moreover, reservoir computing clearly outperforms all approaches for certain tasks like prediction of chaotic time series .?While numerical implementations of this concept exists, any technical implementation is lacking. After the onset of versatile and well-controllable complex dynamics in coupled laser systems has been found and investigated, recently the utilization of nonlinear dynamics for functional purposes has also been discovered. We identified delay-coupled optical systems as ideal substrate for a liquid state machine, since they allow achieving highly complex dynamics and consequently mapping with few elements only.