Parallelizing hierarchical bayesian inference

Hi all, I’m new to numpyro, so this question might be silly. I’ve written a script for hierarchical bayesian inference similar to the one showed in this example. For the purpose of this question, we can assume my code is identical to the one in the example (up to section 3). I have data for thousands of different “patients” (and for each of them I have 11 data points taken at different times). MCMC is running pretty slowly over the complete dataset, so I was wondering if it is possible to separate my data in batches of patients, and run them in parallel in different cores. How should I modify the code to do this? Thanks a lot