Traditional hard metals usually use Co metal as the binder for metal carbides (WC, TiC, VC, etc.), achieving balanced combinations of properties like hardness, toughness, ductility, wear resistance. Co/WC composite has been extensively studied and widely applied in cutting tools, mining and construction, metal forming, wear parts, etc. The primary reasons for choosing Co as the binder material are related to its excellent wetting and sintering properties, a prior combination of toughness, ductility, friction and wear resistance, high thermal stability and so on. However, driven by considerations including health and environmental concerns (e.g., Cobalt is classified as a probable human carcinogen by regulatory agencies such as the International Agency for Research on Cancer (IARC)), cost, supply chain stability, and deficient corrosion resistance in certain environments, academies and industries are exploring various materials as potential alternative binders. High-entropy alloys (HEAs), composed of multi-principal elements, are a promising class of alloy for substituting Co as the binder phase. Designing suitable HEA compositions as the binder phase is however a challenging task. Previous studies are mostly based on the trial-and-error method, and only a small number of HEAs (usually the equiatomic HEAs) have been tested. In the current EU project CoBRAIN, we systematically characterize the key thermodynamic, physical and mechanical properties of HEAs in the large compositional space, and design the binder phase to have the similar properties of Co binder in terms of phase stability, deformation behaviors and magnetic properties, etc. Density functional theory (DFT) calculations have been adopted to study the composition and magnetic dependent phase stability and stacking fault energy, which are critical material properties affecting the transformation-induced plasticity (TRIP) and twinning-induced plasticity (TWIP) effects in the binder phase. Several potential compositions in various HEA systems are identified and further tested as binder in thermal spray coatings to verify the design concept. DFT results, together with experimental and theoretical results from the project, form also as a solid database for developing a fast machine learning method, supporting further optimization and customization of binder phases.