Coinbase stated that it is optimizing the rule creation process in its anti-fraud system by integrating machine learning models and a rule engine to achieve more efficient risk management. It also proposed a dual-track strategy: "models handle long-term defense, rules handle rapid response," and built a unified framework to create a feedback loop: rules are used to capture new types of fraudulent behavior and inversely train the model, thereby continuously improving overall defense capabilities. Specifically, Coinbase has significantly improved efficiency by restructuring data structures, automating schema evolution, and introducing notebook-based analytics tools, transforming the previously manual rule creation process into a data-driven and automated recommendation process. Rule backtesting performance has improved by more than 10 times, and overall response time has been reduced from days to hours. Furthermore, the new system uses machine learning to recommend parameters, helping to reduce false positives and minimizing the impact on legitimate users while combating fraud. Coinbase stated that its next step will be to advance event-driven automated rule generation and explore the "one-click conversion" of efficient rules into model features, further moving towards an automated risk management system.