Blood clots play a crucial role in cardiovascular health. Healthy blood clots form around a vascular injury site to reduce bleeding. In diseased states, clots can obstruct blood flow in major vessels and may break off and cause a stroke. Physiologically realistic blood clots are soft structures with arbitrary shape and heterogeneous porous microstructure. Specifically, blood clots are known to contract during the clotting process. Clot contraction plays a crucial role in both healthy and diseased clots. Clot contraction in healthy clot reduces clot porosity and form a tight seal around the injury site. In a diseased clot, impaired clot contractions reduce clot mechanical rigidity and are shown to correlate with stroke severity [1]. Despite the importance of blood clot contraction mechanisms, comprehensive study of this phenomenon is limited. Complexity of clot structure, interplay with local blood flow, and the underlying biochemical reactions and transport renders considerable challenges in understanding macroscale clot contraction mechanics. Furthermore, computational modeling of blood clot contraction and mechanics using a continuum model can be challenging. The nature of evolving microstructure due to contraction makes modeling the constitutive relation of blood clot unclear. Unlike a continuum representation, discrete particle representation of blood clots enables simulation of porous heterogeneous clots that can fracture under load. In this work, we describe the development of a mesoscopic discrete particle method used to study blood clot contraction and underlying mechanics. We demonstrate the use of this particle-based method to study blood clot rigidity and microstructure under varying degrees of clot contraction, representative of common scenarios in thrombosis and hemostasis.