<p>Achieving consensus with minimal communication overhead and computational complexity remains a core challenge in distributed computing, especially under asynchronous and failure-prone conditions. This paper proposes a scalable and wait-free randomized consensus algorithm based on XOR-based secret sharing for shared coin flipping with a fixed agreement parameter. The protocol executes in two communication rounds: in the first, each process generates a random bit and broadcasts it to all others; in the second, processes compute pairwise shared secrets using XOR operations and then determine the consensus value through parity checks. The use of simple bitwise operations and majority voting ensures that all correct processes converge to the same output bit, regardless of initial randomness. We analyze the algorithm’s correctness, fault tolerance, and complexity, demonstrating that it achieves consensus with O(<InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(\hbox {n}^{2}\)</EquationSource> </InlineEquation>) total operations in the absence of faults. Our approach offers a lightweight alternative to traditional shared coin protocols and is well-suited for scalable and fault-tolerant distributed systems.</p>