Motivated by the fact that the interaction between Nipah virus F protein and ephrin-2B receptor is amongst the highest known for any viral protein (reference: https://doi.org/10.1016/j.cell.2025.02.030), this system is taken as reference.First, the residues involved in the interaction between Nipah virus F protien and ephrin-2B receptor are determined computationally by using geometrical constraints (angle, distance...). From here, the list of interacting residues is generated, and used as hotspot residues for RFDiffussion de novo binder generation. Several proteins are designed de novo with different RFDiffussion checkpoints (all alpha, mixed.... etc) to get as much structural diversity as possible. The interaction interfaces of this newly generated proteins are visually inspected, and the most promising ones are selected for the next phase, where ProteinMPNN (with the SolubleMPNN weights, since we want to generate soluble proteins for experimental testings) is used for sequence generation. Different sampling temperature are used, so we can also obtain a rich sequence variability. Sequences of the generated proteins are clustered with MMseqs2 to delete redundant sequences and just get the representative ones. Designs are then validated by using Boltz2, where the extracellular part of the protein (residues 71-602) are modelled. The glycosilations of the receptor described in UniProt are also taken into account for this modelling task. The most promising models are the ones presented here.