HawkEye is a powerful toolkit for monitoring, observability, and machine learning workflow debugging used internally at Meta. It supports recommendation and ranking models for many of Meta's products. Over the past two years, it has achieved significant improvements in the time required to debug production issues. HawkEye includes infrastructure for continuously collecting data related to services and training models, as well as data generation and analysis components for root cause analysis. It supports guided exploration, investigation, and remediation measures for UX workflows. Through a guided exploration interface based on necessary components, HawkEye allows users to effectively investigate and resolve issues.