The evolution of binary neutron star mergers is governed by an unknown equation of state that can be constrained by multimessenger observations. Gravitational waves emitted post merger are currently undetectable, so the study of remnant evolution relies on electromagnetic observations such as the afterglow and kilonova. These transients can occur within the same time-frame, leading to an entangled signal. Most previous works attempt to tackle this problem by individually fitting kilonova and afterglow models to the contaminated signal, inferring each of the transients' parameters independently. This talk introduces a different approach, joint model fitting, explaining the details behind it- like how machine learning can be applied to speed up inference, and the advantages that come from it.