Principal Machine Learning Scientist, Drug Discovery Analytics

Revolution Medicines•Redwood City, CA

2d•Hybrid

About The Position

Revolution Medicines is a late-stage clinical oncology company developing novel targeted therapies for patients with RAS-addicted cancers. The company’s R&D pipeline comprises RAS(ON) inhibitors designed to suppress diverse oncogenic variants of RAS proteins. The company’s RAS(ON) inhibitors daraxonrasib (RMC-6236), a RAS(ON) multi-selective inhibitor; elironrasib (RMC-6291), a RAS(ON) G12C-selective inhibitor; zoldonrasib (RMC-9805), a RAS(ON) G12D-selective inhibitor; and RMC-5127, a RAS(ON) G12V-selective inhibitor, are currently in clinical development. As a new member of the Revolution Medicines team, you will join other outstanding professionals in a tireless commitment to patients with cancers harboring mutations in the RAS signaling pathway. The Opportunity: We are seeking a Principal Machine Learning Scientist to lead the development of advanced machine learning approaches that accelerate small-molecule drug discovery. This role sits at the intersection of data science, chemistry, and biology, transforming complex scientific datasets into predictive models that guide target discovery, compound design, and translational hypotheses. Working closely with experimental scientists, the Principal ML Scientist will develop cutting-edge modeling approaches that integrate chemical, biological, and phenotypic data. The successful candidate will play a key role in advancing a data-driven discovery strategy by designing predictive models, deploying innovative algorithms, and translating insights into actionable decisions that improve the speed and success of the discovery of medicines for patients with RAS-driven cancers.

Requirements

PhD in machine learning, computational chemistry, computational biology, computer science, or a related quantitative discipline.
8+ years experience applying machine learning or advanced analytics to scientific problems.
Demonstrated experience working with chemical or biological datasets in drug discovery or related domains.
Strong expertise in: Python-based ML ecosystems (PyTorch, TensorFlow, scikit-learn).
Data analysis and scientific computing (NumPy, Pandas).
Deep learning and representation learning techniques.
Strong understanding of early-stage drug discovery workflows.
Ability to translate biological or chemical questions into computational frameworks and predictive models.
Proven ability to communicate complex computational insights to.
Passion for scientific innovation and a relentless commitment to improving patient outcomes.

Nice To Haves

Proven track record of applying advanced AI/ML approaches (deep learning, generative modeling, structure-based ML) to drug discovery or related life sciences domains.
Experience with cheminformatics or bioinformatics toolkits is highly desirable.
Familiarity with cloud computing and scalable ML workflows is a plus
Ability to work at the interface of computational and experimental science.

Responsibilities

Scientific Leadership: Define and lead machine learning strategies that accelerate early-stage drug discovery.
Identify opportunities where AI and advanced analytics can meaningfully improve scientific decision-making.
Drive the adoption of innovative modeling approaches within multidisciplinary discovery teams.
Model Development: Develop predictive models for: Compound activity, selectivity, ADME/Tox, and developability properties.
Target engagement, mechanism-of-action, and phenotypic datasets.
Apply modern ML techniques such as: Graph neural networks.
Deep learning for molecular representation.
Generative chemistry models.
Active learning frameworks for experimental design.
Cross-Functional Collaboration: Partner with medicinal chemists to guide compound design and optimization.
Work with biologists to interpret complex experimental datasets and generate mechanistic hypotheses.
Collaborate with data scientists and engineers and ML engineers to deploy models into scalable discovery workflows.
Data Integration: Integrate heterogeneous datasets including: Chemical structure and screening data.
Imaging and phenotypic screening data.
Structural biology and molecular simulation outputs.