Enhancing Accuracy and Performance of Precessing Spin EOB Models with NRPy and BOB

As gravitational wave interferometers become more sensitive, it is essential to develop more accurate and efficient waveform models for compact binary mergers. To this end, we present novel strategies to enhance the performance and accuracy of precessing spin waveform models based on the Effective-One-Body (EOB) formulation. We start by implementing our version of the state-of-the-art inspiral model, SEOBNRv5PHM, using an intuitive Python infrastructure documented in Jupyter notebooks. To enhance performance, we leverage our Python package, NRPy, to generate highly optimized C code. The optimized inspiral is smoothly attached to our in-house highly accurate Backwards-One-Body (BOB) merger-ringdown model, a first principles model based on the properties of the final merged remnant. By combining performance enhancements from the post-adiabatic formalism with the accuracy of self-consistent spinning EOB evolution at appropriate time scales, we achieve a robust and efficient implementation. We conduct comprehensive comparisons of performance and accuracy between the traditional SEOBNRv5PHM and TEOBResumS-GIOTTO models and our proposed approximant.