Postdoctoral Fellow: Generative AI for Protein Engineering

Pfizer • Berlin, CT

1d • Hybrid

About The Position

At Pfizer, our purpose is to deliver breakthroughs that transform patients’ lives. At the heart of this mission is our Research and Development organization, where advanced science and cutting‑edge technologies are translated into impactful medicines and vaccines. Across discovery, development, and clinical research, our scientists play a critical role in advancing innovative therapies that improve health outcomes for patients around the world. Within Pfizer Research & Development, the Medicine Design (MD) group is seeking a highly motivated Postdoctoral Scholar to conduct independent and collaborative research in computational protein design and optimization. This postdoctoral position focuses on developing and applying modern computational and machine learning–based approaches to accelerate the discovery of next‑generation biologic therapeutics. The role sits at the intersection of machine learning, protein science, and therapeutic discovery, offering an exceptional opportunity to translate foundational AI research into real‑world drug design applications. The selected postdoctoral fellow will pursue high‑impact, publishable research while working closely with multidisciplinary scientists across Pfizer R&D. In addition, the postdoc will join a cohort of AI‑focused postdoctoral researchers spanning a range of research topics, providing a vibrant community for peer mentorship, collaboration, and exposure to diverse applications of machine learning in drug discovery and development. The program strongly supports professional growth through structured mentorship, scientific publication, and career development.

Requirements

• Ph.D. degree in computational chemistry, physical or biological sciences, chemical engineering, computer science, or related discipline.
• Less than 2 years of post-degree experience.
• Willingness to make a minimum 2-year commitment.
• Candidates advancing to interview stage must provide two letters of recommendation.
• Successful record of scientific accomplishments evidenced by scientific publications and/or presentations with at least 2-3 first-author publication in a peer-reviewed journal.
• Strong familiarity with state-of-the-art protein engineering tools (i.e., RFDiffusion, ProteinMPNN (e.g., flow matching/diffusion models/antibody design/protein design/small molecule design).
• Strong working knowledge of modern protein engineering and generative modeling approaches, such as RFdiffusion, ProteinMPNN, and related diffusion‑ or flow‑based methods applied to protein, antibody, and/or small‑molecule design.
• Solid foundation in protein language models and structure‑aware generative models, with demonstrated application to biological sequence and structural data.
• Hands‑on experience with machine learning and computational biology libraries, including PyTorch (required) and RDKit (required).
• Proficiency in Python programming, with experience using components of the scientific Python ecosystem (e.g., NumPy, SciPy, pandas) for research and model development.
• Familiarity with running compute‑intensive machine learning experiments, including experience with GPU‑accelerated workflows and high‑performance computing environments (e.g., Slurm) or a strong interest in developing these skills.
• Strong intellectual curiosity and enthusiasm for data‑driven research, with the ability to translate novel ideas into well‑designed computational experiments and research prototypes.

Nice To Haves

• Prior experience or demonstrated interest in antibody design or protein engineering.
• Experience fine‑tuning foundation models using small and biased datasets.
• Familiarity with antibody developability properties is highly desirable.
• Experience with reinforcement learning, Bayesian optimization, or other advanced multi-objective optimization techniques.
• Experience with molecular dynamics physics-based simulations
• Experience training and deploying models in cloud environments (e.g., AWS, or GCP), including containers (Docker), orchestration (Kubernetes), and basic MLOps practices (versioning, CI/CD, monitoring).

Responsibilities

• Conduct original research in computational protein engineering, with an emphasis on sequence- and structure-based generative modeling for protein fiducial design (i.e., RFDiffusion, ProteinMPNN).
• Develop and deploy state-of-the-art ML methods for multi-objective, constraint-aware protein optimization, balancing affinity, stability, and developability.
• Apply proprietary computational framework and ML models for protein developability engineering.
• Communicate research findings through manuscripts, conference presentations, and internal seminars.
• Be an active member of a highly interdisciplinary team, collaborate with computational and experimental researchers in a multidisciplinary research environment.

Benefits

• a 401(k) plan with Pfizer Matching Contributions and an additional Pfizer Retirement Savings Contribution
• paid vacation, holiday and personal days
• paid caregiver/parental and medical leave
• health benefits to include medical, prescription drug, dental and vision coverage

What This Job Offers

• Career Level: Entry Level
• Education Level: Ph.D. or professional degree
• Number of Employees: 5,001-10,000 employees