Bayesian nonparametric (BNP) models often offer flexibility benefits over their parametric counterparts since their use of infinite-dimensional parameter spaces reduces the need to make unnecessarily strict modeling assumptions. Examples of BNP applications include the Indian buffet process (IBP) in latent factor modeling and the Gaussian process (GP) in nonlinear regression. This paper approaches the task of dynamic latent factor modeling – that is, characterization of latent factors that influence observed trends over time – by proposing the IBPGP model, a BNP model that incorporates nonparametric methods both in relating latent factors to observed trends and in modeling the latent factors as functions of time. The paper then describes a Markov chain Monte Carlo (MCMC) sampling algorithm to estimate model parameters, and it applies this algorithm to a small, simulated dataset. The inference results indicate the potential for recovery of true parameters but also reveal opportunities for improvement in the sampling paradigm.