Associate Scientist, Postdoctoral Fellow – AI/ML and Agentic AI for Drug Discovery

Merck•South San Francisco, CA

2d•$82,000 - $92,000•Hybrid

About The Position

The Postdoctoral Research Fellow will work in Modeling & Informatics (M&I) within Discovery Chemistry at the company's Research Laboratories in South San Francisco, California. The fellowship is part of a program designed to provide industrial postdoctoral researchers with an academic focus in a commercial environment. The fellow will conduct research in agentic artificial intelligence in an industry setting and gain experience in drug discovery. They will have access to high-performance computing resources and internal data assets, collaborate with scientists across research disciplines, and apply modern AI/ML methods to automate and accelerate predictive modeling workflows. The role also offers opportunities to contribute to scientific publications and conference presentations. The fellow will benchmark AI/ML methods on internal and external drug discovery datasets, lead an independent project designing and evaluating AI agent frameworks for drug discovery modeling, and refine workflows for transparency, robustness, and decision-making. They will analyze complex drug discovery datasets and communicate findings through internal forums, conferences, and peer-reviewed journals. Candidates must hold a Ph.D. or expect to receive one no later than spring 2027. Relevant doctoral disciplines include Machine Learning, Computer Science, Computational Chemistry, Cheminformatics, Chemical or Biomedical Engineering, Applied Mathematics, or a related field. The role is hybrid, offers domestic relocation, and provides visa sponsorship. The salary range is $82,000–$92,000, with eligibility for an annual bonus and long-term incentive, if applicable. Benefits include medical, dental, vision, and other insurance benefits for employees and families, retirement benefits including a 401(k), paid holidays, vacation, and compassionate and sick days.

Requirements

Currently hold a Ph.D. or receive a Ph.D. no later than spring 2027.
Ph.D. in Machine Learning, Computer Science, Computational Chemistry, Cheminformatics, Chemical or Biomedical Engineering, Applied Mathematics, or a related discipline.
Demonstrated expertise in machine learning, cheminformatics, or computational chemistry.
Strong record of scientific achievement, evidenced by peer-reviewed publications and presentations.
Hands-on experience building, training, and rigorously evaluating machine learning models in Python.
Experience designing, conducting, and interpreting computational experiments, including model benchmarking.
Strong quantitative and analytical skills, including statistical analysis and data interpretation.
Ability to work independently and thrive in a highly collaborative, multidisciplinary environment.
Excellent oral and written communication skills.

Nice To Haves

Experience with large language models and agentic AI, such as LangGraph or LangChain.
Proficiency with modern ML frameworks and experiment-tracking tools, such as PyTorch and MLflow.
Experience applying graph neural networks, chemical foundation models, or other modern deep learning approaches to molecular or drug discovery data.
Experience with predictive modeling in drug discovery.
Experience with cheminformatics tools such as RDKit.
Experience developing or improving agentic AI systems through feedback, adaptation, or alignment techniques.
Demonstrated creativity and scientific independence in developing and executing research projects.

Responsibilities

Design, execute, and interpret research benchmarking standard AI/ML methods across internal and external drug discovery datasets.
Develop and drive an independent research project to design and evaluate AI agent frameworks for drug discovery modeling.
Refine agentic AI workflows to improve transparency, robustness, and decision-making.
Analyze complex drug discovery datasets using appropriate statistical and machine learning tools.
Present research findings at internal scientific forums, external conferences, and in peer-reviewed scientific journals.