Musicians rely on nuanced microrhythm, small variations in timing, dynamics, and other aspects, to create an expressive rhythmic feel in music performance. Electronic music production often attempts to replicate these qualities through algorithmic manipulations to achieve similar effects. In this work, we address the generation of microrhythm using a method that learns microtiming and dynamics from onset timing and strength annotations of drum performances. We frame microrhythm learning as a sequence modeling task, leveraging a Transformer-based model. Our focus is on Uruguayan candombe drumming, where we explore its rhythmic patterns at both the beat and rhythmic cycle levels. To evaluate the model’s effectiveness in replicating the original microrhythm, we compare the mean, standard deviation and histogram intersection of timing deviations and dynamics values at each subdivision for the original and the generated data. The model is deployed as a VST enabling artists to incorporate candombe grooves into drum scores. With this work, we aim to bridge the gap between algorithmic rhythm creation and the expressive qualities of live performance, striving to produce music with the authentic grooves of various Latin American genres.