Developing structural and machine learning based methods for molecular design within the Research and Early Development organization.
Requirements
Demonstrated experience with machine learning libraries in production-ready workflows (e.g., PyTorch + Lightning + Weights and Biases)
Experience with physical modeling methods (e.g., molecular dynamics) and cheminformatics toolkits (e.g., rdkit)
Previous focus on one or more of these areas: molecular property prediction, computational chemistry, de novo drug design, medicinal chemistry, small molecule design, self-supervised learning, geometric deep learning, Bayesian optimization, probabilistic modeling, statistical methods.
Public portfolio of computational projects (available on e.g. GitHub).
Responsibilities
Manage projects deploying new techniques for machine learning based molecular optimization for the analysis and design of small and large molecule drugs within target-driven design campaigns.
Engineering pipelines for probabilistic molecular property prediction and Bayesian acquisition for active learning based drug discovery.
Engineering pipelines for molecular generative modeling.
Develop machine learning and Bayesian optimization workflows to analyze existing, and design new, small and large molecules.
Work on existing projects and generate new project ideas.
Other
PhD degree in a quantitative field (e.g., Computer Science, Chemistry, Chemical Engineering, Computational Biology, Physics), or MS degree and 3+ years of industry experience.
Record of achievement, including at least one high-impact first author publication or equivalent.
Excellent written, visual, and oral communication and collaboration skills.
You will closely collaborate with scientists across the organization
You will be expected to form close working relationships with small molecule and protein therapeutic development efforts across the organization.
