We have introduced a new strategy, called Genetic Analytic Continuation (GAC), to face inverse problems and we have applied it to the extraction of information about real time dynamics of a many-body quantum system from noisy imaginary time correlation functions f(t) computed via Quantum Monte Carlo (QMC). Production and falsification of model spectral functions s(w) are obtained via a survival-to-compatibility with f(t) evolutionary process, based on Genetic Algorithms. Statistical uncertainty in f(t) is promoted to be an asset via a sampling of equivalent f(t) within the noise, which give rise to independent evolutionary processes, allowing to capture physical properties of s(w) through an averaging strategy. With this method we have studied bulk 4He at T = 0 K; for the first time we recover from exact QMC simulations sharp quasi-particle excitations and we find excellent agreement with experimental data, with spectral functions displaying also the multiphonon branch. The ability of GAC in resolving the single quasi-particle contribution form the multiphonon one, allows for the extraction of the strength of the single quasi-particle peak, Z(q), and from the momenta of the spectral functions s(w) we have also access to quantities like the static density response function X(q); Z(q) and X(q) turn out to be in good agreement with experimental data. We have addressed also the study of one 3He impurity in liquid 4He obtaining the first result for the impuriton branch from exact QMC simulations. We also studied vacancy-waves excitations in hcp solid 4He finding a novel roton like feature along the GM direction. The study of roton excitations in the metastable overpressurized liquid 4He is under way.