Abstract: Over the last decade, there has been significant progress in the field of machine learning-based de novo drug discovery, particularly in generative modeling of chemical structures. However, current generative approaches exhibit a significant challenge: they do not ensure the synthetic accessibility nor provide the synthetic routes of the proposed small molecules which limits their applicability. In this work, we propose a novel reinforcement learning (RL) setup for drug discovery that addresses this challenge by embedding the concept of synthetic accessibility directly into the de novo compound design system. In this setup, the agent learns to navigate through the immense synthetically accessible chemical space by subjecting initial commercially available molecules to valid chemical reactions at every time step of the iterative virtual synthesis process. The proposed environment for drug discovery provides a highly challenging test-bed for RL algorithms owing to the large state space and high-dimensional continuous action space with hierarchical actions. Our end-to-end approach achieves state-of-the-art performance when compared against other generative approaches for drug discovery. Moreover, we leverage our approach in a proof-of-concept that mimics the drug discovery process by generating novel HIV drug candidates. Finally, we describe how the end-to-end training conceptualized in this study represents an important paradigm in radically expanding the synthesizable chemical space and automating the drug discovery process.