Large-scale production of graphene nanosheets (GNSs) has led to the availability of solution-processable GNSs on the commercial scale. The controlled vacuum filtration method is a scalable process for the preparation of wafer-scale films of GNSs, which can be used for gas sensing applications. Here, we demonstrate the use of this deposition method to produce functional gas sensors, using a chemiresistor structure from GNS solution-based techniques. The GNS suspension was prepared by liquid-phase exfoliation (LPE) and transferred to a polyvinylidene fluoride (PVDF) membrane. The effect of non-covalent functionalization with Co-porphyrin and Fe-phthalocyanines on the sensor properties was studied. The pristine and functionalized GNS films were characterized using different techniques such as Raman spectroscopy, scanning electron microscopy (SEM), transmission electron microscopy (TEM), atomic force microscopy (AFM), X-ray diffraction (XRD), and electrical characterizations. The morphological and spectroscopic analyses both confirm that the molecules (Co-porphyrin and Fe-phthalocyanine) were successfully adsorbed onto the GNSs surface through π-π interactions. The chemiresistive sensor response of functionalized GNSs toward the low concentrations of nitrogen dioxide (NO2) (0.5–2 ppm) was studied and compared with those of the film of pristine GNSs. The tests on the sensing performance clearly showed sensitivity to a low concentration of NO2 (5 ppm). Furthermore, the chemical modification of GNSs significantly improves NO2 sensing performance compared to the pristine GNSs. The sensor response can be modulated by the type of adsorbed molecules. Indeed, Co-Por exhibited negative responsiveness (the response of Co-Por-GNS sensors and pristine GNS devices was 13.1% and 15.6%, respectively, after exposure to 0.5 ppm of NO2). Meanwhile, Fe-Phc-GNSs induced the opposite behavior resulting in an increase in the sensor response (the sensitivity was 8.3% and 7.8% of Fe-Phc-GNSs and pristine GNSs, respectively, at 0.5 ppm NO2 gas).