I'd be interesting in getting pymc4 connected with [DifferentialEquations.jl](https://diffeq.sciml.ai/stable/) through [diffeqpy](https://github.com/SciML/diffeqpy) in a way to avoid in a high performance way ([like we did with R](https://www.stochasticlifestyle.com/gpu-accelerated-ode-solving-in-r-with-julia-the-language-of-libraries/)). What is the right way to get started for defining operations with derivative rules? The library [already has a ton of options for defining the derivatives](https://diffeq.sciml.ai/stable/analysis/sensitivity/) so I just need to figure out how to hook the systems together. Thanks for any help getting started.