Machine learning has had a significant impact on the field of accelerator science in recent years, with the number of machine learning based papers presented at accelerator conferences increasing year on year. The CLARA accelerator is currently undergoing commissioning for operation at 250MeV at 100Hz. Virtual diagnostics, anomaly detection, and control systems using machine learning will play critical roles in exploiting the accelerator. In this seminar I’ll be presenting the work we’ve done on machine learning at Daresbury, giving a brief overview of the strategies and challenges, and discussing future research directions.