Machine Learning Post Doctorate Scientist(m/w/x)

# Machine Learning Post Doctorate Scientist - Basel **Basel** | **Full time** ### The Position The Large Molecule Drug Discovery group within Prescient Design / (MLDD - Machine Learning for Drug Discovery) in Roche/Genentech seeks exceptional researchers who have a demonstrated research background in machine learning and protein structural biology and design, a passion for independent research and technical problem-solving, and a proven ability to develop and implement ideas from research. The group provides a dynamic and challenging environment for cutting-edge, multidisciplinary research including access to heterogeneous data sources, close links to top academic institutions around the world, as well as internal Genentech Research and Early Development (gRED) and Pharmaceutical Research and Early Development (pRED) partners and research units. Our mission is to develop and apply machine learning methods in designing novel macromolecules. Researchers in this role will develop and apply new deep learning-based methods for large molecule property prediction and de novo generative design. ### The Opportunity * You will join Prescient Design and you will work with a group of talented AI/ML scientists & engineers, structural/computational biologists located in Basel, Switzerland and hosted by Roche. * Participate in cutting-edge research in ML, structural biology, and physics-based modeling applications to drug discovery and design. * Collaborate closely with cross-functional teams and contribute to therapeutic development efforts across gRED & pRED to solve complex problems including developing models to predict antigen-antibody affinity and developability properties and to perform generative design of *de novo* macromolecules. * Refine models and workflows by performing exploratory data analysis, interrogating scientific hypotheses, and rigorous model selection. * Deliver deep learning based software solutions for accelerating drug discovery, design, and therapeutic development. * Develop the team's culture. Write structured, tested, readable and maintainable code. * Contribute to publications and present results at internal and external scientific conferences, workshops, and venues. ### Who You Are * **Ph.D. in Computer Science, Computational Biology, Statistics, Applied Math, Physics, Chemistry**, or related technical field — to be eligible, you must be within the first four years of completing your PhD at the start of the project * Demonstrated experience with Python and deep learning libraries such as Pytorch and/or Jax, TensorFlow * Extensive knowledge of generative diffusion or flow models for statistical physics systems, such as statistical physics, small molecules, peptides, or proteins. * Excellent communication and collaboration skills with intense curiosity to bridge the field of machine learning and physics-based structural modeling * Demonstrated research experience, including at least one first author publication or equivalent at the top machine learning conferences (e.g., ICML, ICLR, NeurIPS, etc.) **Furthermore, preferred** * Prior experience or familiarity with combining generative models with monte carlo sampling. Extensive knowledge of Monte-Carlo algorithms such as HMC or Molecular Dynamics. * Prior experience in extending autograd engines with custom ops. * Experience with kernel learning, ideally Gaussian processes, in a biological context The duration of the project is initially set for two years, with the possibility of extension for a third year. You will be based in Basel, Switzerland. **Roche is an Equal Opportunity Employer.**