Neural machine translation is increasingly being promoted and introduced in the field of translation, but research into its applicability for post-editing by human translators and its integration within existing translation tools is limited. In this study, we compare the quality of SMT and NMT output of the commercially-available interactive and adaptive translation environment Lilt, as well as the translation process of professional translators working with both versions of the tool, their preference for SMT vs. NMT for post-editing, and their attitude towards such an interactive and adaptive translation tool compared to their usual translation environments.