Consumer-resource models are a central framework for understanding the collective behaviour of ecological communities. In their standard formulation, the interaction coefficients between species and resources are quenched random variables. However, real ecosystems continuously adapt their resource consumption strategies through ecological and evolutionary processes. In this work we investigate an adaptive extension of the MacArthur Consumer Resource Model in which the consumer-resource couplings evolve dynamically in response to species and resource abundances. We focus on a simple adaptation mechanism that reinforces interactions associated with abundant species and abundant resources. Integrating out the adaptive variables generates memory kernels. Using Dynamical Mean Field Theory (DMFT), we derive an effective stochastic description of the model in the thermodynamic limit S, Q large.