The Simulation & Modeling Sciences team seeks a postdoctoral researcher interested in applying reinforcement learning to accelerate pharmaceutical drug design. We are looking for applicants with a research background in reinforcement learning, familiarity with organic chemistry, and the ability to translate ideas from theory to practical drug discovery applications.

Our group sits at the intersection of computational, chemical, and biological sciences, providing an environment for multidisciplinary applied research with access to heterogeneous data sources across Pfizer’s discovery and development organization. Additionally, we have close links to top academic institutions around the world as well as with internal partners and research units. The post-doctoral researcher will primarily focus on areas within reinforcement learning—including deep and end-to-end reinforcement learning, inverse reinforcement learning, game theory, and decision theory—but should be broadly interested in machine learning algorithms that can be leveraged to enable faster drug design and optimization

ROLE RESPONSIBILITIES

• Evaluate and apply a variety of modern reinforcement learning algorithms to drug design.
• Leverage Pfizer’s proprietary data, in-house software, and compute infrastructure to develop scalable tools and automated workflows to assist with early-stage drug discovery.
• Write well-documented, tested, modular code individually and collaboratively atop Pfizer’s Python/C++ technical stack within a high-performance scientific computing environment.
• Effectively communicate the advantages and caveats of developed algorithms to technically-diverse internal audiences and foster partnership across medicinal sciences R&D.
• Collaborate with design teams across the medicinal sciences organization to apply reinforcement learning algorithms to active drug discovery projects.
• Design and publish articles in top peer-reviewed journals in the field and deliver scientific and technical presentations at internal and external venues.

BASIC QUALIFICATIONS

• Ph.D. in Computer Science, Computational Chemistry, or a related technical field with thesis work in machine learning
• Undergraduate-level Organic Chemistry
• Well-cited journal publications and presentations/publications at conferences or workshops (such as CVPR, ICCV, ECCV, NIPS, ICML, ICLR, UAI, AISTATS, etc.)
• Programming experience in Python and/or C++
• Experience with Torch and/or TensorFlow

PREFERRED QUALIFICATIONS

• Research experience working applying machine learning to chemistry in industry or academia
• Advanced coursework in organic or medicinal chemistry
• Programming experience with GPUs
• Strong portfolio of open-source software