Senior Specialist, Data Science / Modeling & Informatics

Merck•South San Francisco, CA

1d•$159,600 - $251,200•Onsite

About The Position

The Modeling and Informatics group, an approximately 50-person team of computational chemists and data scientists in the global Discovery Chemistry organization, applies structure-based modeling and predictive science to drug design and development. The organization is seeking a computational scientist for its on-site research laboratories in South San Francisco, California. This role develops algorithms, tools, and workflows that bring together predictive modeling, agentic scientific reasoning, and physics-based insight to evaluate molecular hypotheses and learn from experimental feedback across the Design-Make-Test-Analyze (DMTA) cycle. The successful candidate will apply expertise across AI/ML, computer science, biophysics, and computational chemistry. They will develop and evaluate computational methods, use modern deep learning informed by physical and biophysical understanding, and create agentic workflows that combine models, scientific tools, data, simulations, structural information, SAR, and literature. The role collaborates with medicinal and computational chemistry teams, pilots methods in active drug-discovery programs, contributes to reproducible computational workflows and broader discovery platforms, conducts research, publishes and presents scientific work, and mentors scientists. This is an on-site position involving in-person collaboration at the research laboratories in South San Francisco; it is not remote. The position is regular, with a first-shift/day schedule. The listed salary range for the primary location is $159,600–$251,200, and the successful candidate is eligible for an annual bonus and long-term incentive, if applicable. Relocation is available domestically or internationally, visa sponsorship is available, and travel requirements are 10%.

Requirements

Bachelor’s degree with 7 or more years of relevant experience, Master’s degree with 4 or more years of relevant experience, or a Ph.D. (or Ph.D. completion by June 2027) in Engineering, Computer Science, Machine Learning, Chemistry, Computational Chemistry, Biophysics, Applied Mathematics, or a related discipline.
At least two years of hands-on experience, including experience obtained during education, developing, training, and evaluating deep-learning models with a framework such as PyTorch or JAX.
Broad competency in modern data science.
Research experience applying deep-learning approaches to chemistry or biophysics problems.
Experience with agentic AI frameworks.
Track record of scientific publications in peer-reviewed journals.
Strong Python skills for scientific programming, methods development, and algorithm development on Linux high-performance computing.
Excellent written and oral communication skills.

Nice To Haves

Experience with AI/ML methods relevant to molecular design, such as graph neural networks, transformers, diffusion models, generative modeling, reinforcement learning, active learning, or molecular foundation models.
Experience with agentic AI frameworks such as LangChain, LangGraph, or similar.
Experience with cheminformatics tools such as RDKit.
Experience with 3D molecular modeling and physics-based computational methods such as protein-ligand modeling, structure-based molecular design, molecular dynamics, or quantum chemistry.
Scientific software-engineering experience translating research algorithms into robust, reproducible computational workflows.
Experience collaborating directly with experimental scientists.

Responsibilities

Develop AI/ML algorithms and computational methods for molecular representation, prediction, generation, reasoning, and decision-making, including components for AI co-scientist and autonomous molecular-design systems.
Evaluate AI/ML methods using appropriate computational and experimental benchmarks, assessing scientific validity, uncertainty, robustness, baselines, failure modes, and prospective utility.
Develop and apply deep-learning approaches to molecular discovery that combine machine learning with physical and biophysical understanding of molecular systems.
Develop agentic workflows that integrate predictive models, scientific tools, experimental data, simulations, structural information, SAR, and scientific literature to generate and evaluate molecular hypotheses.
Collaborate with medicinal and computational chemistry project teams on molecular design, SAR interpretation, hypothesis generation, and experimental prioritization.
Build reproducible scientific methods and workflows, pilot approaches on active drug-discovery programs, and collaborate on deploying successful methods across computational environments and broader discovery platforms.
Conduct research in computational chemistry, computer science, biophysical chemistry, and molecular modeling relevant to drug discovery; contribute through publications and presentations.
Mentor scientists and help expand organizational capabilities in data-driven molecular design, computational methodology, and scientific reasoning.