CasADi is a versatile open-source framework for numerical optimization in general and numerical optimal control in particular. Over the past 10 years, it has been used for numerous applications in academia and industry. In this tutorial, we provide a hands-on demonstration of the new FMI for Model Exchange interoperability in CasADi. Topics covered include the calculation of validated first and second order derivatives, forward and adjoint sensitivity analysis, finding steady-state solutions, and numerical optimal control. Bring your laptop for the exercises, offered in Python and Matlab.