The investigation and control of liquid flow at the nanometer scale is a key area of applied research with high relevance to physics, chemistry, and biology. We introduce a method and a device that allows the spatial resolution of liquid flow by integrating an array of graphene-based magnetic (Hall) sensors that is used for tracking the movement of magnetic nanoparticles immersed in a liquid under investigation. With a novel device concept based on standard integration processes and experimentally verified material parameters, we numerically simulate the performance of a single sensor pixel, as well as the whole sensor array, for tracking magnetic nanoparticles having typical properties. The results demonstrate that the device enables (a) the detection of individual nanoparticles in the liquid with high accuracy and (b) the reconstruction of a particle's flow-driven trajectory across the integrated sensor array with sub-pixel precision as a function of time, in what we call the "Magnetic nanoparticle velocimetry" technique. Since the method does not rely on optical detection, potential lab-on-chip applications include particle tracking and flow analysis in opaque media at the sub-micron scale.