Information technology requires more and more high-performing devices for information encoding and processing. In this regard the use of optical solitons as information bits appears promising, especially if implemented in fast, compact and cheap devices as semiconductor lasers. In particular, "light-bullets" (LB), where the light would be localized in the three dimensions of space, are expected to lead to disruptive performances in terms of bit rate, resilience and agility. The aim of this project is to conceive, to realize and to operate semiconductor laser devices for the generation and control of spatiotemporal solitons, also called “light bullets” (LB). LB will be implemented in Vertical External-Cavity Surface Emitting semiconductor devices mounted in an external cavity configuration (VECSEL) closed by a saturable absorber mirror (SESAM). Devices fabrication will be developed in the frame of this project to match the parameters requirements for LB existence. Once LS will be obtained and characterized, their application to information processing will be addressed by targeting a three-dimensional all-optical buffer. LB have been chased in conservative systems since the pioneer work by Silberberg at the beginning of ’90. The propagation of an optical pulse in a medium where diffraction and anomalous group dispersion are both compensated by a non-linearity is strongly unstable and, despite the efforts made, it is impossible to avoid the pulse to collapse or to spread. The originality of our approach to LB consists in implementing them in dissipative system, where LB will appear as stable solutions for a wide set of initial conditions and control parameters. In addition, when the system is strongly dissipative, LB can be individually addressed by an external (optical) perturbation and used as information bits. More precisely, LB we are aiming at in this project are spatio-temporal “Localized Structures” (LS). LS have been observed in the transverse section (spatial LS) and in the longitudinal direction (temporal LS) of optical resonators. Several experiments have disclosed the potential of LS for information processing, especially when implemented in fast and scalable media as semiconductor resonators. LB we will obtain will lead to three-dimensional buffering of data inside the VECSEL external cavity. If the transverse section of the device allows creating an array of NXN spatial bits and the longitudinal cavity allows for storing M bits, one may handle MXNXN bits in a single device by using LB as information bits. The temporal bit rate is accordingly increased by a factor given by NXN with respect to single-transverse mode resonators. The performances obtained in past experiments in semiconductor lasers lead to an estimation of 5 Kbit sequences stored in the cavity and a writing/reading bit rate of 100 GS/s. Beyond information processing, LB are very interesting for other applications where picoseconds laser pulses are required at an arbitrary low repetition rate and at an arbitrary pattern sequence (time-resolved spectroscopy, optical code division multiple access communication networks and LIDAR). The possibility of integrating metasurfaces onto the VECSEL or onto the SESAM will induce vorticity to each light bullet, thus enabling the creation of an array of optical tweezers for parallel manipulation of biological nano-objects. The use of semiconductor lasers for supporting LB is an important aspect of our project. If implementation of LB in semiconductor lasers enhances their attractiveness for applications, the conception and manufacturing of devices able to sustain these structures is challenging and a large part of the project will be devoted to devices optimisation.