We describe Lapedo, a novel library of hybrid parallel skeletons for programming heterogeneous multi-core/many-core CPU/GPU systems in Erlang. Lapedo's skeletons comprise a mixture of CPU and GPU components, allowing skeletons to be flexibly and dynamically mapped to available resources, with all of the low-level tedious code to divide work between CPUs and GPUs, transfer the data between the main and GPU memory and offload computations to the GPUs provided by the library. We evaluate the effectiveness of Lapedo on three realistic use cases from different domains, demonstrating significant improvements in speedups compared to CPU-only and GPU-only executions.