We propose a realization of an information machine which converts information to work, using the concept of Leo Szilard's feedback protocol. In our realization, a colloidal particle is drifting with fluid flow in a 1D channel until it encounters a barrier. The particle is monitored as it diffuses. Once it moves further enough upstream, the barrier is moved closer to the particle without affecting it. Repeating this feedback based action results in transporting the particle itself. We aim to implement this model both in experiment and in simulation. Our experimental design is comprised of a colloidal particle diffusing in a microfluidic channel, with a repelling laser based barrier. Here, we present preliminary results regarding the feasibility of such experiments. On the simulation front, we have implemented a Langevin dynamics simulation and studied the relevent time scales of the system's dynamics. We have also studied the effect of sampling and feedback cycles on the efficiency of the information machine.