Although there has been for the past 20 years great interest in the synthesis and use of metal nanoparticles, little attention has been paid to the complexity of the surface of these species. In particular, the different aspects concerning the ligands present, their location, their mode of binding, and their dynamics have been little studied. Our group has started in the early 1990s an investigation of the surface coordination chemistry of ruthenium and platinum nanoparticles but at that time with a lack of adequate techniques to fulfill our ambition. Over 10 years later, we went back to this problem and could obtain a more precise vision of the surface species. This Account is centered on ruthenium chemistry. This metal has been the most studied in our group, first thanks to the availability of a precursor, Ru(cyclooctadiene)(cyclooctatriene) (Ru(COD)(COT)), which possesses the ability to decompose in very mild conditions without leaving residues on the resulting nanoparticles and second because of the absence of magnetic perturbations (Knight shift, paramagnetism, ferromagnetism, etc.), which has allowed the use of solution and solid state NMR. In this respect, it has been possible to evidence the presence of a high concentration of hydrides on the surface of these particles, to study their dynamics, and to show that since the polarity of the Ru-H bond is similar to that of the C-H bond, a Ru/H NP would behave as a big lipophilic entity. The second point was to characterize the coordination of ancillary ligands. This has been achieved for different ligands, in particular phosphines and carbenes, which made possible the study of the modification of NP reactivity induced by surface ligands. This led to the conclusion that the presence of surface ligands can benefit both the activity of NP catalysts and their selectivity. If it was expected that the selectivity could be modulated, the promoting effect from the presence of ligands on, for example, arene or CO hydrogenation was totally unexpected. Playing with poison atoms (Sn, Fe, etc.) or ligands (CO) may allow us to play with the reactivity of the NPs to make them more selective for selected reactions. Finally, the search for specific ligands for nanoparticles is still in its infancy, but some examples have been found as have specific reactions of nanoparticles. Obviously arene hydrogenation and CO hydrogenation were well-known in heterogeneous catalysis, but we could demonstrate that they can be carried out in very mild conditions on ligand stabilized RuNPs. On the other hand, the enantiospecific C-H activation leading to enantioselective labeling of large organic or biomolecules or the C-C bond cleavage in mild conditions were both unexpected. There is still much work to perform for reaching the degree of control on nanoparticles that is presently achieved in organometallic molecular chemistry, but this work shows that it is possible.