Kubernetes has become the de-facto platform for hosting container-based applications and is one of the most rapidly adopted technologies in the industry. Despite its widespread use, Kubernetes remains a complex system, particularly for individuals who are new to it or lack deep technical expertise. An informal survey conducted by StackState in January 2023, involving about 100 practitioners, revealed that over two-thirds of participants struggled with troubleshooting Kubernetes. Integrating artificial intelligence (AI) and machine learning into the Kubernetes workflow is becoming increasingly important in addressing these challenges. In this paper, I present an in-depth exploration of K8sGPT, an open-source, AI-driven tool that leverages large language models (LLMs) to assist Operations, DevOps teams, and developers in identifying, diagnosing, and proposing relevant solutions to Kubernetes issues. We will conduct an experiment to demonstrate the capabilities of K8sGPT by deploying a broken application and observing how the tool facilitates troubleshooting. K8sGPT, being open-source and vendor-agnostic, can be deployed in both on-premises and cloud environments, making it a versatile solution for enhancing Kubernetes operations. This paper aims to showcase how K8sGPT can transform the troubleshooting process, making it more efficient and accessible for users at all levels of expertise.