Context. Binarity and multiplicity appear to be a common outcome in star formation. In particular, the binary fraction of massive (OB-type) stars can be very high. In many cases, the further stellar evolution of these stars is affected by binary interactions at some stage during their lifetime. The origin of this high binarity and the binary parameters are poorly understood because observational constraints are scarce, which is predominantly due to a dearth of known young massive binary systems. Aims. We aim to identify and describe massive young binary systems in order to fill in the gaps of our knowledge of primordial binarity of massive stars, which is crucial for our understanding of massive star formation. Methods. We observed the two massive young stellar objects (MYSOs) PDS 27 and PDS 37 at the highest spatial resolution provided by VLTI/PIONIER in the H-band (1.3 mas). We applied geometrical models to fit the observed squared visibilities and closure phases. In addition, we performed a radial velocity analysis using published VLT/FORS2 spectropolarimetric and VLT/X-shooter spectroscopic observations. Results. Our findings suggest binary companions for both objects at 12 mas (30 au) for PDS 27 and at 22–28 mas (42–54 au) for PDS 37. This means that they are among the closest MYSO binaries resolved to date. Conclusions. Our data spatially resolve PDS 27 and PDS 37 for the first time, revealing two of the closest and most massive (>8 M_⊙) YSO binary candidates to date. PDS 27 and PDS 37 are rare but great laboratories to quantitatively inform and test the theories on formation of such systems.