The past decades have seen the advent of information theory in various fields, from quantum physics to cosmology. At an intermediary scale between atomic and cosmological scales are biological systems and in particular the cell, as a constitutive element of any living organism. Molecular mechanisms in the cell are tasked to perpetually create and maintain order, against the natural evolution suggested by the laws of thermodynamics. Many agree that these mechanisms are strongly related to the processing of information, but so far most of the established parallels remain at the stage of conceptual analogy. This leaves an obvious gap in a formal and rigorous de- scription of cellular active processes: all the elements are in place at the biochemical level to process information. Through the internal structure of the involved macromolecules, biologi- cal systems are able to autonomously evolve, as a result of various information-processing steps. To this end, the present work articulates around two examples of cellular transport, in order to highlight the molecular mechanisms inherent to the processing of information. In the first part, we establish a model for the transport by ABC Transporters and identify the conditions required for the creation of a concentration gradient across the membrane. The emerging conditions are explicit demonstrations of the different steps in the transport of substrate during which information is processed, induced by the structural and biochemical properties of the transporter. The conclusions extend way beyond a simple conceptual analogy: ABC transporters are autonomous Maxwell Demons. In the second part, our work focuses on the dynamics of Hsp70-driven substrate translocation through membrane. Our model reveals that optimality lies in a balance between the strength of the directionality and the intrinsic diffusion rate of the substrate through the pore. These numerical observations pave the way to a new experimental exploration: the oligomerization of Hsp70 might accelerate the translocation, as an adaptative response to imposed conditions in the surroundings. Finally, we show how kinetic properties of chaperone proteins emerge from the resolution of a simple and minimal model for translocation. Such a model also displays a mathematical structure which brings us to state that the translocation machinery is an autonomous Szilard engine, continuously processing information. Both projects converge to a bridge between information theory and non-equilibrium thermo- dynamics applied to active cellular transporters. The analogy between information-processing devices and biological molecular systems is not only conceptual: active cellular mechanisms evolved to perpetually and autonomously process information. This results in the creation and maintainance of order, that acts as a cornerstone of life.