Troubleshooting remains largely a manual process, with substantial time devoted to diagnosing issues. This session explores how AI/ML can expedite initial diagnoses. It delves into the suitability of various AI/ML techniques for network data and different phases of analysis, such as problem identification and diagnosis. Using real-world telemetry data as a foundation, we will collectively examine and demonstrate the use of diverse AI/ML approaches, including large language models, to pinpoint and diagnose issues. Concurrently, we will discuss the advantages and drawbacks of the various technologies employed.