Large-scale semidefinite programming has many applications, including optimal control, computer vision, and machine learning. However, current algorithms for solving semidefinite programs (SDPs) can be time consuming and memory intensive. We look at new heuristics for solutions of the K disjoint clique problem. We model the K disjoint clique optimization problem as a SDP based on non-convex low rank factorization, and solve using Alternating Direction Method of Multipliers, augmented Lagrangian, and alternating direction. We will present numerical results illustrating the efficacy of our approach for clustering of real and simulated data and pose future questions of interest.