Binary black hole GW200129 is claimed to be the highest precessing binary ever observed. In fact, the measured orbital precession is about 10 orders of magnitude higher than any previous measurement. However, a broadband noise disturbance in LIGO Livingston overlapped the gravitational-wave signal, putting the precession claims in doubt. In this talk, I will present a state-of-the-art neural network that is able to model and mitigate the broadband noise from the LIGO Livingston interferometer. I will also show that our neural network mitigates the noise better than the algorithm used by the LIGO-Virgo collaboration. In the end, I will revisit GW200129 precession claims by presenting re-analysed data using our noise-mitigated frames.