Senior Scientist, Computational Biotherapeutics Engineering

Use Your Power for Purpose

At Pfizer, our purpose is to deliver breakthroughs that transform patients' lives. Central to this mission is our Research and Development team, which strives to convert advanced science and cutting-edge technologies into impactful therapies and vaccines. Whether you are engaged in discovery sciences, ensuring drug safety and efficacy, or supporting clinical trials, your role is crucial. You will leverage innovative design and process development capabilities to expedite the delivery of top-tier medicines to patients globally.

What You Will Achieve

Pfizer’s BioMedicine Design department is seeking a scientist with expertise in computational AI/ML methods spanning protein modeling, representation learning, and generative model development.  You will join a computational team working closely with experimental scientists on discovery and optimization of industry‑leading biotherapeutics. You will implement, evaluate, and apply state‑of‑the‑art AI/ML methods to advance biotherapeutic discovery and engineering, integrating new models into scalable discovery workflows and decision‑making. By collaborating across departments, you will shape the next generation of AI/ML architectures, training strategies, and evaluation approaches for biotherapeutic design.  This role offers the opportunity to directly influence the design of clinical molecules, helping achieve Pfizer’s goal of breakthroughs that change patients’ lives.

How You Will Achieve It

• Implement advanced cutting-edge AI and machine learning workflows for computational protein design (including fine-tuning protein language models and generative protein design) in HPC or scalable cloud computing environments

• Collaborate with machine learning colleagues on the design and training of AI/ML models for antibody developability engineering.  Apply these models to optimize leads for antibody drug discovery projects.

• Stay informed about developments in NLP, ML, and generative AI to create innovative solutions for molecular discovery, design, and optimization to advance therapeutic discovery and development.

• Serve as a technical expert in deep learning models for protein sequence and structure, supporting discovery teams with AI/ML‑driven design strategies.

• Analyze large‑scale sequence, structure, and experimental datasets to learn representations linking protein features to developability and pharmaceutical properties.

• Communicate complex scientific ideas, model behavior, limitations, and design recommendations to both technical and non-technical audiences, fostering collaborations across multidisciplinary teams.

• Collaborate with computational and wet lab experts to optimize the computational developability platform, offering both individual and team-based innovative solutions.

Qualifications

Must‑Have

• PhD in biochemistry, computational chemistry, computational biology, machine learning, or a related field with 0 t 3 years of experience OR Master's Degree in biochemistry, computational chemistry, computational biology, machine learning with 7 to 8 years of experience OR BA/BS with 9 to 11 years of experience

• Demonstrated track record (including publications or equivalent impact) of using AI/ML‑driven protein modeling/design to influence project direction and strategy

• Hands‑on experience using and interrogating modern AI/ML models for protein representation, structure prediction, or generation (e.g., transformer or diffusion-based approaches).

• Strong understanding of protein structure, sequence–structure relationships, and model evaluation.

• ·Experience programming in Python and using modern scientific or machine learning libraries (e.g., NumPy/SciPy, scikit-learn, PyTorch), including training and evaluation workflows.

• Experience working with large biological datasets and bioinformatics resources.

Nice to Have

• Experience with protein language models (e.g., ESM‑family models), generative structure models (e.g., RFdiffusion, BoltzGen, BindCraft), and structural prediction AI models (e.g. AlphaFold)

• Familiarity with equivariant or structure-aware neural networks.

• Knowledge of antibody structure, multispecific design, or developability modeling.

• Experience running and scaling deep learning workloads on HPC/GPU/cloud environments using technologies such as Slurm, AWS, or Google cloud.

• Experience with structure-based molecular modeling software (Rosetta, Schrödinger, MOE, FoldX)

  
Work Location Assignment: Hybrid

Hybrid colleagues