This article describes my research on the discovery of innovative small-molecule targeted therapeutics for use in personalized, precision cancer medicine. This is exemplified by two particular examples. Firstly, I describe our discovery of novel and effective small-molecule inhibitors of the molecular chaperone HSP90 – and especially the leading clinical agent, AUY922, which is showing strong activity in HER2-positive breast cancers that have become refractory to trastuzumab (Herceptin®, Roche Ltd, Basel, Switzerland) and also in non-small cell lung cancers, including those that are resistant to oncogenic driver protein kinase inhibitors – with general potential to overcome or prevent the problem of drug resistance. Secondly, I discuss our discovery of pathfinding small-molecule inhibitors of class I PI3 kinases, especially the first-in-class pan-class I-selective inhibitor GDC-0941, for which clinical responses have been observed in breast and ovarian cancer, gastrointestinal stromal tumours, multiple myeloma and melanoma. More broadly, our Cancer Research UK Cancer Therapeutics Unit has discovered 17 innovative drug candidates since 2005 and progressed seven of these into biomarker-led, hypothesis-driven clinical trials with successful pharmacodynamic biomarker modulation in all cases and clinical responses in most to date, exemplifying the use of our ‘Pharmacologic Audit Trail’ concept for drug discovery and clinical development. Furthermore, our CYP17 inhibitor abiraterone (Zytiga®, Janssen Pharmaceuticals Inc., NJ, USA), designed and synthesized in our unit and initially trialled in our partner hospital, The Royal Marsden Hospital (London, UK), was approved for late-stage castration-resistant prostate cancer in 2011. Thus, we have developed a successful model for high-quality academic drug discovery in the non-profit setting (Anonymous. Drug discovery gets an academic push. Cancer Discov 2011; 1:93–4). Importantly, this provides a creative derisking solution for the innovation gap or ‘valley of death' that often divides basic research from clinical translation. This new model is especially critical at a time when the adverse economic climate and other pressures have led to downsizing and increased risk aversion in the large pharmaceutical companies together with reduced venture investment in biotechnology companies. Having taken on the early high-risk research of new targets, we have also been very successful in partnering our drug discovery projects with industry at the appropriate stage using a range of commercial models with speed-to-patient benefit as the main driver [Houlton S. Fast, flexible and flourishing. Chemistry World, 2012. URL: http://www.rsc.org/chemistryworld/more/?author=35&date-range=2-5+years (last accessed 30 September 2014)]. Before focusing on our work on what I refer to as ‘drugging the cancer genome’, and particularly the examples of HSP90 and PI3 kinase inhibitors, I provide a short summary of the background to molecularly targeted cancer drug discovery and development together with brief comments on my early research on hypoxia-targeted drugs and signal transduction inhibitors, with the aim of setting the context for what follows. To finish, I offer some concluding remarks and future perspectives on the discovery, development and use of personalized, precision medicines for cancer.