One of the possible approaches for solving a CSP is to encode the input problem into a CNF formula, and then use a SAT solver to solve it. The main advantage of this technique is that it allows to benefit from the practical efficiency of modern SAT solvers, based on the CDCL architecture. However, the reasoning power of SAT solvers is somehow "weak", as it is limited by that of the resolution proof system they use internally. This observation led to the development of so called pseudo-Boolean (PB) solvers, that implement the stronger cutting planes proof system, along with many of the solving techniques inherited from SAT solvers. Additionally, PB solvers can natively reason on PB constraints, i.e., linear equalities or inequalities over Boolean variables. These constraints are more succinct than clauses, so that a single PB constraint can represent exponentially many clauses. In this paper, we leverage both this succinctness and the reasoning power of PB solvers to solve CSPs by designing PB encodings for different common constraints, and feeding them into PB solvers to compare their performance with that of existing CP solvers.