Accurate sleep stage classification is crucial for diagnosing sleep disorders and understanding sleep patterns. Traditional supervised learning methods require large amounts of labeled data, which is often expensive and time-consuming to obtain. In this work, we explore a self-supervised approach for sleep stage classification using the SimCLR framework combined with contrastive learning techniques. By leveraging unlabeled sleep recordings, our method learns robust feature representations that can be fine-tuned with a limited set of labeled examples. We design augmentations suitable for physiological signals and adapt the contrastive loss to capture the temporal dynamics of sleep data. Experimental results demonstrate that our approach achieves competitive performance compared to fully supervised baselines while significantly reducing the reliance on labeled data. This study highlights the potential of contrastive self-supervised learning for efficient and scalable sleep stage classification.