An endovascular guidewire manipulation is essential for minimally-invasive clinical applications. This paper introduces a scalable learning pipeline to train AI-based agent models toward automated endovascular predictive device controls. Specifically, we create a scalable environment by pre-processing 3D CTA images, providing patient-specific 3D vessel geometry and the centerline of the coronary. Next, we apply a large quantity of randomly generated motion sequences from the proximal end to generate wire states associated with each environment using a physics-based device simulator. Then, we reformulate the control problem to a sequence-to-sequence learning problem, in which we use a Transformer-based model, trained to handle non-linear sequential forward/inverse transition functions. We then present the safety ratio and difference between the estimated force and the ground-truth in the test set. Our AI-based agents provide an efficient approach to indicate when to turn on/off X-ray.