Strong statistical fluctuations in meso- and nano-scale structures make their thermodynamic properties extremely dependent on the information available about them. The most basic process illustrating the importance of information to statistical systems is the information-to-energy conversion in the famous Maxwell's Demon (MD). Our primary goal is to study both experimentally and theoretically the statistics of fluctuations and the role of information in thermodynamics of the nano-scale systems. The first milestone will be the experimental realization of the nanoscale MD. We will create an experimental set-up and develop the corresponding theory of the monitored statistical evolution with feedback that optimizes the information-to-energy conversion. Our vision is to develop the nanoelectronic and bio-molecular devices that will allow us to systematically explore the limits of information-powered systems, in particular to test the Szilárd's limit relating one bit of information to extracted energy. We will also study statistics of energy fluctuations as revealed via equilibrium and non-equilibrium fluctuations of temperature. Part of these fluctuations has a quantum mechanical origin, but identification of this contribution in practice poses a challenging problem. Another novel extension of the MD work will be the study of thermodynamic constraints on quantum detectors. The principal novelty of our project is that it brings a rigorous experimental component to the field presently dominated by theory. Though the concept of a MD is tremendously important for development of modern statistical mechanics, MD-type experiments are still at their infancy. Our experimental study of MD will naturally lead to further progress in the relevant theoretical concepts.